NETWORK ANALYSIS

AND SYNTHESIS
FRANKLIN

F.

KUO

SECOND EDITION
1.50

i-

->

PETER CLARKE
SECOND HAND BOOKS
Tl.

Rochdale 50514

Network
Anal/sis and

Synthesis
Second Edition

Wiley International Edition

Network
Analysis and
Synthesis
Second Edition

by Franklin
Bell

F.

Kuo

Telephone Laboratories,

John Wiley

&

Sons,

Inc.,

New York

Toppan Company. Ltd.

London

Inc.

Sydney

Tokyo, Japan

1962, 1966 by John Wiley

Copyright
All Rights Reserved

& Sons, Inc.

This book or any part thereof

must not be reproduced in any form
without the written permission of the publisher

Wiley International Edition

This book
to

which

it

not to be sold outside the country

is consigned by the publisher.

is

Libray of Congress Catalog Card

Number

66-16127

Printed in Singapore by Toppan Printing Co. (S) Pte. Ltd.

To My

Father and Mother

Preface

In the second edition, I have tried to keep the organization

of the first
Most of the new material are additions aimed at strengthening
the weaknesses of the original edition. Some specific
changes deserve
mention. The most important of these is a new chapter
edition.

on computer
computers have
brought about many significant changes in the content of
engineering
subject matter concerned with both analysis and
design. In analysis,
computation has become an important adjunct to theory. Theory
estate
lishes the foundation of the subject matter; computation
provides clarity,
depth, and insight. In design, the computer has not
only contributed
precision and speed to existing procedures but has
made practicable
design methods that employ iteration and simulation. The
importance of
computer-aided design cannot be overemphasized. In Chapter 15
I have
attempted to survey some digital computer applications in the
areas of
network analysis and design. I strongly encourage all students
applications (Chapter 15).

In the past

five years, digital

to read

this chapter for cultural interest, if

It

not for survival.

Another new section contains a rigorous treatment of the unit impulse.

was difficult to decide whether to incorporate this material in Chapter
2

in the discussion of signals or in a separate appendix.

ized functions in

By putting

general-

an appendix, I have left the decision of whether to teach

the rigorous treatment up to the individual instructor.
Other changes worth mentioning are: (1) two new sections
on the
Fourier integral in Chapter 3 (2) a section on initial and
final conditions
in Chapter 5; (3) a section on Bode plots in Chapter
8; (4) revised
material on two-port parameters in Chapter 9; and
(5) new sections on
frequency and transient responses of filters in Chapter 13. Major
or
minor changes may be found in every chapter, with the exception
of
Chapters 11 and 12. In addition, many new problems are included
at the
end of each chapter.
;

vii

Preface

viii

Chapters 1 and 2
brief description of the subject matter follows.
characteristics of
general
certain
deal with signal representation and
and includes the
analysis,
Fourier
linear systems. Chapter 3 deals with

and 5
impulse method for evaluating Fourier coefficients. Chapters 4
domain.
time
the
in
equations
differential
discuss solutions of network
two preceding
In Chapters 6 and 7, the goals are the same as those of the
domain.
frequency
the
of
that
here
is
viewpoint
chapters, except that the
system function.
Chapter 8 deals with the amplitude, phase, and delay of a
Chapter
synthesis.
network
with
concerned
are
chapters
The final seven
realizability theory
of
elements
the
10,
Chapter
In
two-ports.
9 deals with
elementary drivingare presented. Chapters 11 and 12 are concerned with
Chapter 13, some
In
procedures.
synthesis
function
transfer
and
point
Chapter 14
fundamental concepts in modern filter design are introduced.
design.
and
analysis
deals with the use of scattering matrices in network
digital
of
survey
brief
And, as mentioned earlier, Chapter 15 contains a
five
are
there
addition,
computer techniques in system analysis. In
functions,

appendices covering the rudiments of matrix algebra, generalized

complex variables, proofs of Brune's theorems, and a visual aid to

filter

approximation.
intended for a two-semester course in network theory.
or
Chapters 1 through 8 may be used in a one-semester undergraduate
analysis.
system
linear
or
analysis
transient
beginning graduate course in

The book

is

network
Chapters 9 and 15 are to be used in a subsequent course on
synthesis.

The second edition is largely a result of the feedback from the professors
who have used this book and from their students, Who have discovered
express my sincere
errors and weaknesses in the original edition. I wish to

who provided this feedback.

due to the following people: Robert Barnard of
are
thanks
Special
Belove and Peter Dorato of the Polytechnic
Charles
State,
Wayne
Sherman of the Bell
Institute of Brooklyn, James Kaiser and Philip
Telephone Laboratories, Evan Moustakas of San Jose State College,
Welch of the University of Texas, and David Landgrebe of Purdue.

appreciation to those

A.
I

J.

am particularly indebted to Mac Van Valkenburg of Princeton University

and Robert Tracey of Illinois for editorial advice and to Donald Ford
of Wiley for help and encouragement.
Zicchino, and
addition, I wish to thank Elizabeth Jenkins, Lynn
In

Joanne Mangione of the Bell Telephone Laboratories for their

and careful typing of the manuscript.
F. F.

tj
u.
Heights,
Berkeley

New Jersey,
December, 1965

efficient

Kuo

Preface to the First Edition

This book is an introduction to the study of electric networks based upon

a system theoretic approach. In contrast to many present textbooks,
the emphasis is not on the form and structure of a network but rather on
excitation-response properties. In other words, the major theme is
concerned with how a linear network behaves as a signal processor.
Special emphasis is given to the descriptions of a linear network by its
system function in the frequency domain and its impulse response in the
time domain. With the use of the system function as a unifying link, the
its

transition

from network

analysis to synthesis can

be accomplished with

relative ease.

The book was

originally conceived as a set of notes for a second course

in network analysis at the Polytechnic Institute of Brooklyn. It assumes

had a course in steady-state circuit analysis.

should be familiar with Kirchhoff's laws, mesh and node equations,
standard network theorems, and, preferably, he should have an elementary
that the student has already

He

understanding of network topology.

brief description of the subject matter follows.

Chapters

and 2

deal with signal representation and certain characteristics of linear net3, 4, 5, and 6 discuss transient analysis from both a time
domain viewpoint, i.e., in terms of differential equations and the impulse
response, and a frequency domain viewpoint using Fourier and Laplace
transforms. Chapter 7 is concerned with the use of poles and zeros in
both transient and steady-state analysis. Chapter 8 contains a classical

works. Chapters

treatment of network functions.

The final five chapters deal with network synthesis. In Chapter 9, the
elements of realizability theory are presented. Chapters 10 and 1 1 are
concerned with elementary driving-point and transfer function synthesis
procedures. In Chapter 12, some fundamental concepts in modern filter
design are introduced. Chapter 13 deals with the use of scattering matrices
ix

Preface to the First Edition

in network analysis

and

synthesis.

In addition, there are three appendices

and proofs
covering the rudiments of matrix algebra, complex variables,
of Brune's realizability theorems.
The book is intended for a two-semester course in network theory.
or
Chapters 1 through 7 can be used in a one-semester undergraduate
analysis.
system
beginning graduate course in transient analysis or linear
Chapters 8 through 13 are to be used in a subsequent course on network
synthesis.

was my very good fortune to have studied under Professor M. E. Van

Valkenburg at the University of Illinois. I have been profoundly influenced
emphasis
by his philosophy of teaching and writing, which places strong
It

of exposition. In keeping with this philosophy, I have tried

viewpoint, and I have
to present complicated material from a simple
exercises. In addition,
and
examples
illustrative
included a large number of
rigor and
mathematical
between
ground
middle
I have tried to take a

upon

clarity

Unless a proof contributes materially to the

it is omitted in favor of an intuitive argument.
theorem,
of
a
understanding
of unit impulses, a development in terms
treatment
in
the
example,
For

intuitive understanding.

unit
of a generalized function is first introduced. It is stressed that the
whose
functions
of
sequence
a
actually
but
function
a
not
really
is
impulse
notion of an
limit point is undefined. Then, the less rigorous, intuitive
the
impulse "function" is presented. The treatment then proceeds along

nonrigorous path.

There are a number of topics which have been omitted. One of these
at present. I have puris network topology, which seems to be in vogue
that
posely omitted topology because it seems out of place in a book
analysis.
network
de-emphasizes the form and structure approach to
In an expository book of this nature, it is almost impossible to reference
field of
adequately all the original contributors in the vast and fertile

network theory. I apologize to those whose names were omitted either

through oversight or ignorance. At the end of the book, some supplementary textbooks are listed for the student who either wishes to fill in
some gaps in his training or wants to obtain a different point of view.
given to me by my
I acknowledge with gratitude the help and advice
former colleagues
by
my
and
Laboratories
Telephone
Bell
colleagues at the
my sincere
express
wish
to
I
Brooklyn.
of
Institute
Polytechnic
at the
were
appreciation to the many reviewers whose advice and criticism
invaluable in revising preliminary drafts of the manuscript. Professors
Stanford
R. D. Barnard of Wayne State University and R. W. Newcomb of
entire
of
the
University deserve specific thanks for their critical reading

manuscript and numerous helpful suggestions and comments.

Miss Elizabeth
In addition, I wish to thank Mrs. Elizabeth Jenkins and

Preface to the First Edition

La

xi

Jeunesse of the Bell Telephone Laboratories for their efficient and

careful typing of the manuscript.

Finally, to my wife Dora, I owe a special debt of gratitude. Her
encouragement and cooperation made the writing of this book an

enjoyable undertaking.

Murray

Hill,

January, 1962

New Jersey,

F. F.

KUO

Contents
Chapter

I:

Signals and Systems

1.1

Signal Analysis

1.2

Complex Frequency
Network Analysis
Network Synthesis

1.3

1.4

Chapter 2:

Signals

14

and Waveforms

20

General Characteristics of Signals

General Descriptions of Signals
The Step Function and Associated Waveforms
The Unit Impulse

20
24
28
33

The Frequency Domain: Fourier Analysis

46

3.1

Introduction

3.2

Orthogonal Functions
Approximation Using Orthogonal Functions

46
47
48
50
52

2.1

2.2
2.3

2.4

Chapter 3:

3.3

3.4
3.5

3.6

Fourier Series
Evaluation of Fourier Coefficients
Evaluation of Fourier Coefficients Using Unit
Impulses

3.7

The Fourier

3.8

Properties of Fourier Transforms

Integral

xiii

58
63
67

xiv

Contents

Chapter 4:

Differential Equations

75

4.1

Introduction

75

4.2

Homogeneous Linear Differential Equations

Nonhomogeneous Equations

76

4.4

Step and Impulse Response

82
85

4.5

Integrodifferential Equations

91

4.6

Simultaneous Differential Equations

93

4.3

Chapter 5:

Network

Analysis:

100

100
103
106

5.1

Introduction

5.2

5.5

Network Elements
Initial and Final Conditions
Step and Impulse Response
Solution of Network Equations

5.6

Analysis of Transformers

114
122

The Laplace Transform

134

6.2

The Philosophy of Transform Methods

The Laplace Transform

6.3

Properties of Laplace Transforms

134
135
137

6.4

Uses of Laplace Transforms

Partial-Fraction Expansions
Poles and Zeros
Evaluation of Residues
The Initial and Final Value Theorems

5.3

5.4

Chapter 6:
6.1

6.5

6.6
6.7
6.8

Chapter

7:

7.1

7.2
7.3

7.4
7.5

7.6

Transform Methods

in

Network Analysis

The Transformed Circuit

Thevenin's and Norton's Theorems
The System Function
The Step and Impulse Responses
The Convolution Integral
The Duhamel Superposition Integral

HI

144
148
155

162
165

175

175
180

187
194
197
201

Contents

Chapter 8:

212

8.1

Amplitude and Phase Response

212

8.2

Bode

221

Plots

8.3

Single-Tuned Circuits

8.4

Double-Tuned Circuits
On Poles and Zeros and Time Delay

229
238
245

Network

253

8.5

Chapter 9:

Analysis:

II

9.1

Network Functions

9.2

Relationships Between Two-Port Parameters

Transfer Functions Using Two-Port Parameters

9.4

Interconnection of Two-Ports

9.5

Incidental Dissipation

9.6

Analysis of Ladder Networks

253
264
266
271
276
279

10:

Elements of Readability Theory

290

9.3

Chapter

Amplitude, Phase, and Delay

xv

10.1

Causality and Stability

10.2

Hurwitz Polynomials
Positive Real Functions
Elementary Synthesis Procedures

290
294
299
308

Synthesis of One-Port Networks with Two Kinds

of Elements

315

10.3

10.4

Chapter

11:

11.1

11.2
11.3

11.4
11.5

11.6

Properties of L-C Immittance Functions

Synthesis of L-C Driving-Point Immittances
Properties of R-C Driving-Point Impedances
Synthesis of R-C Impedances or R-L. Admittances
Properties of R-L Impedances and R-C Admittances

Synthesis of Certain

R-L-C Functions

315
319
325
329
331
333

xvi

Contents

Elements of Transfer Function Synthesis

341

12.1

Properties of Transfer Functions

341

12.2

Zeros of Transmission
Synthesis of Yn and ZS1 with a 1-Q Termination
Synthesis of Constant-Resistance Networks

345
347
352

Topics in Filter Design

365

The Filter Design Problem

The Approximation Problem in Network Theory
The Maximally Flat Low-Pass Filter Approximation

365
365
368
373
388
392

Chapter

12:

12.3
12.4

Chapter

13:

13.1

13.2
13.3

13.4

Other Low-Pass Filter Approximations

13.5

Transient Response of Low-Pass Filters

13.6
13.7

A Method to Reduce Overshoot in Filters

A Maximally Flat Delay and Controllable Magnitude

Approximation
Synthesis of Low-Pass Filters
13.9 Magnitude and Frequency Normalization
13.10 Frequency Transformations
13.8

Chapter

14:

395
397

402
404

413

The Scattering Matrix

14.1

Incident and Reflected

14.2
14.3

The
The

14.4

Properties of the Scattering Matrix

14.5

Insertion Loss

413
415
419
426
429

14.6

Darlington's Insertion Loss Filter Synthesis

431

Chapter

15:

Power Flow

Network
a Two-Port Network

Scattering Parameters for a One-Port

Scattering Matrix for

Computer Techniques

in Circuit Analysis

in Circuit Analysis

15.1

The Uses of Digital Computers

15.2

Amplitude and Phase Subroutine

A Fortran Program for the Analysis of Ladder
Networks
Programs that Aid in Darlington Filter Synthesis

15.3

15.4

438

438
450
453
457

Contents

Appendix A:

xvii

Introduction to Matrix Algebra

461

Fundamental Operations
Elementary Concepts
Operations on Matrices
Solutions of Linear Equations

461

A.2
A.3
A.4
A. 5

References

A.1

Appendix B:

462
464
468
469

on Matrix Algebra

Generalized Functions and the Unit Impulse

470

B.l

Generalized Functions

B.2

Properties of the Unit Impulse

470
476

Elements of Complex Variables

481

C.l

Elementary Definitions and Operations

C.2
C.3
C.4

Analysis

481
483
486
487

Appendix C:

Appendix D:

Appendix E:

Singularities and Residues

Contour Integration

Proofs of
Functions

An Aid

Some Theorems on

Positive

Real
490

to the Improvement of Filter Approxi-

mation

493

493
494
495
496
498
502
504

E.l

Introduction

E.2
E.3
E.4

Constant Logarithmic Gain Contours

Constant Phase Contours

Contour Drawings

E.5

Correction Procedure

E.6

Correction Network Design

E.7

Conclusion

Bibliography

05

Name Index

$09

Subject Index

si I

chapter

Signals and systems

This book is an introduction to electric network theory. The first half

of the book is devoted to network analysis and the remainder to network
synthesis and design. What are network analysis and synthesis? In a
generally accepted definition of network analysis

and

synthesis, there are

three key words: the excitation, the network, and the response as depicted
in Fig. 1.1. Network analysis is concerned with determining the response,
given the excitation and the network. In network synthesis, the problem
to design the network given the excitation and the desired response.
In this chapter we will outline some of the problems to be encountered
in this book without going into the actual details of the problems.

is

We

will also discuss

I.I

some

basic definitions.

SIGNAL ANALYSIS

networks, the excitation and response are given in terms of

and currents which are functions of time, t. In general, these
functions of time are called signals. In describing signals, we use the two
universal languages of electrical engineering time and frequency. Strictly
speaking, a signal is a function of time. However, the signal can be

For

electric

voltages

described equally well in terms of spectral or frequency information. As

between any two languages, such as French and German, translation is
needed to render information given in one language comprehensible in the

Response

Excitation

Network

>-

FIG.

I.I.

The

>

objects of our concern.

Network

analysis and synthesis

m
Ao-

FIG.

Sinusoidal signal.

1.2.

is effected by the
and the Laplace transform. We shall
have ample opportunity to define and study these terms later in the book.
At the moment, let us examine how a signal can be described in terms of
both frequency and time. Consider the sinusoidal signal

other.

Between time and frequency, the translation

Fourier series, the Fourier integral,

s(t)

where

is

the amplitude,

=A
O is

sin (a>

the phase

0.1)

O)

shift,

and

<w

is

the angular

frequency as given by the equation

coo

where

<

L2>

T is the period of the sinusoid. The signal is plotted aguinst time

An equally complete description of the signal is obtained if we

in Fig. 1.2.

aD Ao
io.
E

b>0

Angular frequency

FIG.

1.3a. Plot

of amplitude

versus angular frequency

o>o

to

Angular frequency

FIG. 1.3b. Plot of phase 6 versus angular frequency

<o.

<o.

Signals and systems

6>4 6)3

FIG.

a>2 6>i

&>0

1.4a. Discrete

fc>4 W3 ft)2

FIG.

let

the angular frequency

signal

is

is

is

+<<>

ft>4

0)2

0>3

ft>4

+0)

phase spectrum.

be the independent variable. In

described in terms of ^4

amplitude

ft>i

1.4b. Discrete

co

"3

amplitude spectrum.

b)i

ft)

">1

>

w o>

anl

plotted against frequency,

^0.

as

and in

shown in

Fig. 1.36,

this case, the

where
where phase shift
Fig. 1.3a,

plotted.

Now suppose that the signal is made up of 2n +

s(t)

=%A
<

sin

(<v

sinusoidal components

(1.3)

fy)

The spectral description of the signal would then contain 2n + l lines at

( as given in Figs. 1.4a and b. These discrete spectra
<*>i, >2
of amplitude A versus co and phase shift 6 versus co are sometimes called
line spectra. Consider the case when the number of these spectral lines
become infinite and the intervals oo i+1 w f between the lines approach
zero. Then there is no longer any discrimination between one frequency
and another, so that the discrete line spectra fuse into a continuous spectra,
as shown by the example in Figs. 1.5a and b. In the continuous case, the
sum in Eq. 1.3 becomes an integral

xo
where A{m)

is

known

A(o>) sin [mt

0(to)]

as the amplitude spectrum

dm
and

(1.4)

0(co)

as the phase

spectrum.

As we

shall see later, periodic signals

such as the sine wave in Fig. 1.2

can be described in terms of discrete spectra through the use of Fourier

series. On the other hand, a nonperiodic signal such as the triangular

Network

anal/sis

and synthesis
Mo>)

<a

FIG.

1.5a.

FIG.

Continuous amplitude spectrum.

1.5b.

Continuous phase spectrum.

pulse in Fig. 1.6 can only be described in terms of continuous spectra

through the Fourier integral transform.

1.2

COMPLEX FREQUENCY

In this section,

we

shall see, the

we

will consider the concept of

complex frequency. As

complex frequency variable

s

= a + ja>

(1.5)

a generalized frequency variable whose real part a describes growth and

decay of the amplitudes of signals, and whose imaginary part/cu is angular
frequency in the usual sense. The idea of complex frequency is developed
by examining the cisoidal signal
is

S(f)

FIG.

1.6.

aJ*
Ae

Triangular signal.

(1.6)

Signals and systems

ReS

FIG.

1.7.

Rotating phasor.

when S(0 is represented as a rotating phasor, 1 as shown in Fig. 1.7.

The angular frequency m of the phasor can then be thought of as a velocity
end of the phasor. In particular the velocity to is always at right
shown in Fig. 1.7. However, consider the general
case when the velocity is inclined at any arbitrary angle y> as given in
Figs. 1.8a and 1.86. In this case, if the velocity is given by the symbol s,
we see that s is composed of a component m at right angle to the phasor S
as well as a component a, which is parallel to S. In Fig. 1.8a, s has a
component a toward the origin. As the phasor S spins in a counterclockwise fashion, the phasor decreases in amplitude. The resulting wave
for the real and imaginary parts of S(r) are damped sinusoids as given by
at the

angles to the phasor, as

ReS(f)

ImS(0

= Ae~" cos cot

= X<r*'sina>f

(1.7)

]<b)

FIG. 1.8. (a) Rotating phasor with exponentially decreasing amplitude,

phasor with exponentially increasing amplitude.
1

(b)

Rotating

A phasor S is a complex number characterized by a magnitude and a phase angle

(see

Appendix C).

Network

analysis

and synthesis

/Envelope = Ae~

FIG.

1.9.

Damped

sinusoids.

/ ^~ Envelope = Ae**

FIG.

1. 10.

Exponentially increasing sinusoid.

Signals and systems

FIG.

which are shown in

I.I I.

Fig. 1.9.

Exponential signals.

Note that the damped sinusoid has an

In Fig. 1 .84, the phasor is shown with

exponential envelope decay, Ae~" %

component of velocity +o. Therefore, as the phasor spins,

the amplitudes of the real and imaginary parts increase exponentially
with an envelope Ae+at as shown by Im S(f) in Fig. 1.10.
a positive

real

From

this discussion, it is

S(0

apparent that the generalized cisoidal signal

= Ae" = Ae {a+Mt

(1.8)

and decay of the amplitudes in addition to angular

When a = 0, the sinusoid is undamped, and
the signal is an exponential signal

describes the growth

frequency in the usual sense.

wheny'a>

0,

= Ae*'
(1.9)
Finally, if a = ja> = 0, then the signal is a constant
S(0

as

shown in Fig.

A. Thus

1.3

we

1.11.

see the versatility of a

complex frequency

description.

NETWORK ANALYSIS

As mentioned before, the characterization of the excitation and response

and frequency makes up only part of the analysis problem.
The other part consists of characterizing the network itself in terms of
signals in time

time and frequency, and determining how the network behaves as a signal
processer. Let us turn our attention now to a brief study of the properties
of linear networks and the general characteristics of signal processing by

a linear system.

Network

analysis

and synthesis

BASIC DEFINITIONS
Linear

system (network) is linear if (a) the principle of superposition and

of proportionality hold.

(b) the principle

By

the superposition principle,

[eg(0, r 2(f)l
e[t)

if,

for a given network, [ex (t), r^t)]

and

are excitation-response pairs, then if the excitation were

= ex(t) + e2(t),

would be

the response

r(t)

= r,(f) + ri).

By

the

proportionality principle, if the excitation were C^it), where

x is a
constant, then the response would be C^r^t), i.e., the constant of propor-

Cx

preserved by the linear network. The two conditions of

summarized in Fig. 1.12.
Another definition of a linear network is that the excitation and response
of the network are related by a linear differential equation. We shall
discuss this definition in Chapter 4 on differential equations.

tionality

is

linearity are

Passive

A linear network is passive* if (a) the energy delivered to the network is

nonnegative for any arbitrary excitation, and (b) if no voltages or currents
appear between any two terminals before an excitation is applied.

Reciprocal

A network is said to be reciprocal if when the points of excitation and

measurement of response are interchanged, the relationship between

Cieift)

C2 e2(t)

..

System

System

CuiM

C2 r^t)

Cm(t) *

Ci*t(0+Ck*gt)
System

FIG.

1.12.

CW<)

linear system.

1
G. Raisbeck, "A Definition of Passive Linear Networks in Terms of Time and
Energy," /. Appl. Phys., 25 (Dec. 1954), 1510-1514.

Signals and systems

tit)
r(t)

>-

System

.M.

3<

Ht-Tit

-Ti)

7\

TihTi

Tl

FIG.

excitation

System

1.13.

Time-invariant system.

and response remains the same, Thus must be true

for any

choice of points of excitation and response.

Causal

We say a system is causal if its response is nonanticipatory, i.e., if

=
KO =

(0
then

In other words, a system is causal

oo
/
T, the response is zero for

Time

<T

'

<r

(1.10)

if

before an excitation

< <
t

is

applied at

T.

invariant

A system is time invariant ife(t) -* rtf) implies that e(t

T)-*-r(t T),
where the symbol - means "gives rise to." To understand the concept
of time invariance in a linear system, let us suppose that initially the
excitation is introduced at f = 0, which gives rise to a response r(t). If
the excitation were introduced at / = T, and if the shape of the response
waveform were the same as in the first case, but delayed by a time T
(Fig. 1.13), then we could say the system is time invariant. Another way
of looking at this concept is through the fact that time-invariant systems
contain only elements that do not vary with time. It should be mentioned
here that linear systems need not be time invariant.
Derivative property

From the time-invariant property we can show that,

if e{f)

at the input

gives rise to r(i) at the output (Fig. 1.14), then, if the input were

e'(f),

Network

10

analysis

and synthesis

System

de(t)

System

>
*

dr(t)

System

d*Ht)

'

dt'

it

d*e(t)

dt*

FIG.

1.14.

Some

k
>

System

e(T)dr

dt

C'.{,\
l'(T)
J
1

implications of linear time-invariant systems.

r'(t).
The proof is quite
where e is a real quantity. By the
time-invariant property, the response would be r(t + e). Now suppose the

i.e.,

the derivative of e(t), the response would be

simple. Consider an excitation e(*

e)

excitation were

e1 (r)--W*

)-)]

(1-11)

then according to the linearity and time-invariant properties, the response

would be

= - [r(I + e) -

ri (0

Taking the limit as

e ->- 0,

we

K0]

(112)

see that

limc1(0
e-0

= 7-c(0
at
(1.13)

lim rx(0
-o

= ^ K0
at

We

can extend this idea to higher derivatives as well as for the integrals of

e(t)

and

Ideal

r(f),

as

shown

in Fig. 1.14.

models

Let us now examine some idealized models of linear systems. The

systems given in the following all have properties which make them very
useful in signal processing.

Signals and systems

Km.

e(t)

Amplifier

FIG.

1.15. Amplifier.

fit)

K dtd
K
FIG.

>K i
FIG.

Amplifier:

1.

An

1.18.

Time-delay network.

1.19.

2. Differentiator:

amplifier scales

The input

Excitation function.

up

= Ke(t), where Kis a constant (Fig.

up or down

ff VaT

/Tt-JX

uyif

lit)

dt

1.17. Integ[rat or.

FIG.

Kdfltt

.16. Differentiator.

FIG.

the magnitude of the input,

i.e.,

1.15).

signal is differentiated

and possibly scaled

(Fig. 1.16).

3. Integrator:

The output

is

the integral of the input, as

shown

in

Fig. 1.17.
4. Time delayer: The output is delayed by an amount
same wave shape as the input (Fig. 1.18).

Suppose we take the triangular pulse in

T,

but retains the

Fig. 1.19 as the input signal.

the outputs for each of the four systems just described are shown
in Figs. 1.200-1 .20(2.

Then

12

Network analysis and

synthesis

m
1

-1
<b)

tit)

_L
Ti

Ti

Ti

(d)

(c)

FIG. I JO. la) Amplifier output,

(ft)

Differentiator output, (c) Integrator output,

(rf)

Delayed output.

Ideal elements

In the analysis of electric networks, we use idealized linear mathematical

models of physical circuit elements. The elements most often encountered
are the resistor R, given in ohms, the capacitor C, given in farads, and
the inductor L, expressed in henrys. The endpoints of the elements are
port is defined as any pair of two terminals into which
called terminals.
energy is supplied or withdrawn or where network variables may be
measured or observed. In Fig. 1.21 we have an example of a two-port

network.
The energy sources that

are ideal

current or voltage sources,

polarities

make up the excitation functions

as shown in Figs. 1.22a and b. The

Two-port
network

FIG.

1.21.

Two-port network.

*a Response

'/ measurement

Signals and systems

13

w Q)

p
FIG. 1.22a. Voltage source.

FIG. 1.22b. Current source.

indicated for the voltage source

and the direction of flow for the current

source are arbitrarily assumed for reference purposes only. An ideal
voltage source is an energy source that provides, at a given port, a voltage
signal that

is

independent of the current at that port. If we interchange the

words "current" and "voltage" in the

last definition,

we then

an

define

ideal current source.

In network analysis, the principal problem is to find the relationships

that exist between the currents and voltages at the ports of the network.
Certain simple voltage-current relationships for the network elements also
serve as defining equations for the elements themselves. For example,
when the currents and voltages are expressed as functions of time, then the

R, L, and

t>(0

C elements,

shown

in Fig. 1.23, are defined

= L -J9

or

i(t)

= -v(t)

or

iQ)

=i

(0

=7

I'
CJo

v(x)

L *o
J

at

(*)

dx

+ v(0)

or

i(t)

=C

by the equations

dx

j(0)

(1.14)

^
dt

where the constants of integration

discussed in detail later.

i'(0)

and

(0) are initial conditions to

be

Expressed as a function of the complex frequency variable s, the equations

m*v(t)

m
v(t)

M
FIG.

Ut)
*
v(t)

(b)

1.23. (a) Resistor,

(b) Inductor,

^=C

(e)

(c) Capacitor.

Network

14

analysis and synthesis

Ks)

1(3)

>

V(s)

]sL

V(s)

1.24. (a) Resistor,

defining the R, L,
initial

(c)

(b)

(a)

FIG.

sC

V(s)

and

conditions for the

(b) Inductor,

shown

elements,

(c)

Capacitor.

in Fig. 1.24, are (ignoring

moment)

V(s)

V(s)

= sLI(s)

RI(s)

or

I(s)

= ^V(s)

or

I(s)

V(s)

(1.15)

sL
7(s)

= -^/(s)

or

/(s)

sCV(s)

sC

We see that in the time domain, i.e., where the independent variable is

t,

the voltage-current relationships are given in terms of differential equations.

On the other hand, in the complex-frequency domain, the voltage-current

elements are expressed in algebraic equations.

Algebraic equations are, in most cases, more easily solved than differential
relationships for the

equations. Herein lies the ration d'itre for describing signals

in the frequency domain as well as in the time domain.

and networks

When

a network is made up of an interconnection of linear circuit

elements, the network is described by its system or transfer function H(s).
The response R(s) and the excitation E(s) are related by the equation
R(s)

= H(s) E(s).

(1.16)

In network analysis, we are given E(s), and we can obtain H(s) directly
from the network. Our task is to determine R(s).

1.4

NETWORK SYNTHESIS

We will now briefly introduce some of the problems germane to network

synthesis.

In network synthesis,

excitation E(s),

we

and the
from the

are given the response R(s)

and we are required to

synthesize the network

Signals and systems

m <b
FIG.
Z{s)

VM

1.25. Driving-point

15

FIG. I.M. Black box.

impedance

R.

system function

aw-

(1.17)

(s)

Since R(s) and (5) are voltages or currents, then H(s) is denoted generally
as an immittance if R(s) is a voltage and E(s) is a current, or vice versa.
9
driving-point immittance is defined to be a function for which the

variables are measured at the

Z(s) at a given port

is

same

Thus a

port.

driving-point impedance

the function

TO-

(1.18)

I(s)

where the excitation

shown

is

in Fig. 1.25.

a current

and the response is a voltage V(s), as

interchange the words "current" and

I(s)

When we

"voltage" in the last definition, we then have a driving-point admittance.

An example of a driving-point impedance is the network in Fig. 1.25,

where
Z(s)

= ^i = R

(1.19)

I(s)

Now suppose the resistor in Fig. 1.25 were enclosed in a "black box."
We have no access to this black box, except at the terminals 1-1' in Fig.
1.26. Our task is to determine the network in the black box. Suppose we
are given the information that, for a given excitation /(*), the voltage
response V(s) is proportional to I(s) by the equation

V(s)

= KI(s)

An obvious solution, though not unique,

is

(1.20)

that the network consists of a

KCl. Suppose next that the excitation

of value R
V(s), the response is a current I(s), and that

resistor

ns)
IRE

_M.

+ 4j

is

a voltage

(1.21)

Standards on Circuits "Linear Passive Networks," Proe. IRE, 48, No. 9

(Sept. 1960), 1608-1610.

Network

16

analysis

and synthesis

Us)

hM

'

lo-

>i

=r 4

Two-port
network

Vi(s)

V&)

l'o
FIG.

1.27.

Network

FIG.

realiza-

1.28.

Two-port network.

tion for Y(s).

a network equivalent to the network in the black

box. From a close scrutiny of the driving-point admittance Y(s), we see
that a possible solution might consist of a resistor of J O in parallel with
a capacitor of 4 farads, as seen in Fig. 1.27.
The problem of driving-point synthesis, as shown from the examples
just given, consists of decomposing a given immittance function into

Our task is to

synthesize

basic recognizable parts (such as 3

+ 4s).

Before

we proceed with

the

mechanics of decomposition, we must first determine whether the function

is realizable, i.e., can it be synthesized in terms of positive resistances,
inductances, and capacitances? It will be shown that realizable drivingpoint immittances belong to a class of functions known as positive real or,
simply, p.r. functions. From the properties of p.r. functions, we can test a
given driving-point function for realizability. (The Appendices present a
short introduction to complex variables as well as the proofs of some
theorems on positive real functions.) With a knowledge of p.r. functions,
we then go on to examine special driving-point functions. These include
functions which can be realized with two kinds of elements only the L-C,

R-C, and R-L immittances.

Next we proceed to the synthesis of transfer functions. According to
4
the IRE Standards on passive linear networks, a transfer function or
transmittance is a system function for which the variables are measured at
different ports. There are many different forms which a transfer function
might take. For example, consider the two-port network in Fig. 1.28.
If the excitation is I^s) and the response Vjs), the transfer function is a
transfer impedance

Ztt(s)

(1.22)

On the other hand, if Vl(s) were the excitation and

we would

)=

tf(S

Fi()
4

Loe.

eit.

V^s) the response, then

have a voltage-ratio transfer function

(1.23)

Signals and systems

17

\H(ju)\

FIG.

As

1.2?. Ideal

amplitude spectrum for low-pass

filter.

for driving-point functions, there are certain properties which a

transfer function

must

satisfy in

these realizability conditions

order to be realizable.

We

shall study

and then proceed to the synthesis of some

simple transfer functions.

The most important aspect of transfer function

synthesis

is filter

design.

A filter is defined as a network which passes

a certain portion of a frequency spectrum and blocks the remainder of the spectrum. By the term
"blocking," we imply that the magnitude response \H(ja>)\ of the filter is
approximately zero for that frequency range. Thus, an ideal low-pass
a network which passes all frequencies up to a cutoff frequency
and blocks all frequencies above w c as shown in Fig. 1.29.
One aspect of filter design is to synthesize the network from the transfer
function H(s). The other aspect deals with the problem of obtaining a

filter is

mc

realizable transmittance H(s) given the specification of, for example, the

magnitude characteristic in Fig.

1.29.

This part of the synthesis is generally

referred to as the approximation problem.

mation?" Because frequency response

Why

the

word "approxiand C

characteristics of the R, L,

elements are continuous (with the exception of isolated points called

made to
we can realize low-pass filters

resonance points), a network containing these elements cannot be

cut off abruptly at

<o c

in Fig. 1.29. Instead,

which have the magnitude

characteristics

of Fig.

1.30.

In connection with

\H(J<*)\

FIG.

1.30. Realizable

low-pass filter characteristics.

Network

18

analysis

and synthesis

we

problems in magnitude
a filter, we deal with
element values such as R 0.5 ohm and C = 2 farads instead of "practical"
element values of, for example, R 500,000 ohms and C = 2 picofarads
(pico = 10-12). Also we will study a method whereby low-pass filter
designs might be transformed into high-pass, band-pass, and bandelimination filters. The mathematical basis of this method is called
the

filter

design problems,

will discuss certain

and frequency normalization so

that, in designing

frequency transformation.
We next discuss some aspects of analysis and synthesis in which the
excitation and response functions are given in terms of power rather than

of voltage and current.

We will examine the power-transfer properties of

which describe the incident

power of the network at its ports.
Finally, in Chapter 15, we will examine some of the many uses of highspeed digital computers in circuit analysis and design. In addition to a
general survey of the field, we will also study some specific computer
programs in circuit analysis.
linear networks, using scattering parameters,

and

reflected

Problems
1.1

Draw the line

s(t)

1.2
JtTir/4

spectra for the signal

3 sin (t

+ ^1 + 4sin

(it

- 1| +

6 sin it

Find the response to the excitation sin / into a sampler that closes every
= 0, 1, 2,
seconds where
Draw the response for ^ / ^ 2*.

13 Find the response to the excitation shown in the figure when the network
is (a)

an

ideal differentiator;

(ft)

an ideal integrator.

Mt)

_L

PROS.
1.4

If the system function of

1.5

1
1.3

a network

is

given as

H{s)~
(s

+ 2X* +

3)

find the response R(s) if the excitation

Signals and systems

19

network

realiz-

is

s
1.5
ations.

Given the driving-point functions find

their simplest

Z(s)

()

3*
(*)

Y(s)

= 2s

Z(s)

id)

r(s)

-3i+

s*+2
2,

3s

s*

+4

1.6 For the network shown, write the mesh equation in terms of (a) differential
equations and (b) the complex-frequency variable s.

R
l
wv
npfir*

-0
PROB.
1.7

1.6

For the network shown, write the node equation in terms of (a) differential

equations and (6) complex-frequency form.

PROB.
1.8
r(t)

1.7

Suppose the response of a linear system to an excitation

(/)

= 3e-*'. What would the response be to an excitation of t(t 2)1

were

chapter

Signals and

Our main concern

in this chapter

is

waveforms

the characterization of signals as

signals
functions of time. In previous studies we have dealt with d-c
of
sinusoids
were
which
signals
a-c
or
time,
with
that were constant
practice,
engineering
In
sin
(cot
A
0).
s(t)
as
such
constant amplitude,
than
the class of signals encountered is substantially broader in scope

of
simple a-c or d-c signals. To attempt to characterize each member
signals
of
variety
infinite
almost
the
of
view
in
foolhardy
is
the class
be
encountered. Instead, we will deal only with those signals that can
building
characterized in simple mathematical terms and which serve as
will concentrate on formublocks for a large number of other signals.
than deal with
lating analytical tools to aid us in describing signals, rather

We

limitathe representation of specific signals. Because of time and space

behavior,
random
exhibit
tions, we will cover only signals which do not
characterized as functions of time.
i.e., signals which can be explicitly

These signals are often referred to as deterministic

signals.

Let us

first

discuss certain qualitative aspects of signals in general.

2.1

GENERAL CHARACTERISTICS OF SIGNALS

In this section we will examine certain behavior patterns of signals.

Once these patterns are established, signals can be classified accordingly,
and some simplifications result. The adjectives which give a general
continuous.
qualitative description of a signal axe periodic, symmetrical, and
Let us discuss these terms in the given order.
signal
First, signals are either periodic or aperiodic. If a
it is

is

periodic, then

described by the equation

s(t)

= s(tkT)

k
20

0,1,2,...

(2.1)

Signals

and waveforms

21

-2T -r

T 2T 3T AT 5T
-l

FIG.

where

2.1.

Square wave.

T is the period of the signal. The sine wave, sin is periodic with
T = lit. Another example of a periodic signal is the square wane
in Fig. 2.1. On the other hand, the signals given in Fig. 2.2 are
/,

period
given

aperiodic, because the pulse patterns

do not repeat

interval T. Alternatively, these signals

may be considered "periodic" with

an

after

a certain

finite

infinite period.

Next, consider the symmetry properties of a signal. The key adjectives

signal function can be even or odd or neither.
here aie even and odd.

An even function obeys the relation

j(0-*(-0
For an odd function
For example, the function
pulse in Fig. 2.2a

is

s(i)

sin

t is

(2.2)

-s{-t)

(2.3)

odd, whereas cos

t is

even, whereas the triangular pulse

even.
is

odd

The square
(Fig. 2.26).

Observe that a signal need not be even or odd. Two examples of signals
of this type are shown in Figs. 2.3a and 2.4a. It is significant to note,
however, that any signal s(t) can be resolved into an even component 5,(0
and an odd component s9(t) such that
*(0

= *.(0 + *o(0

(2.4)

For example, the signals in Figs. 2.3a and 2.4a can be decomposed into
odd and even components, as indicated in Figs. 2.3b, 2.3c, 2.4b, and 2.4c.

m
o

t
2

M
FIG. 12.

(b)

(a)

Even

function, (b)

Odd function.

22

Network

analysis

and synthesis
s(t)

1
i

-1

-t

(a)

(a)

sjt)

sM

-1

i
1

(b)

(b)

soft)

oW

-1

-i
(0

(O
FIG.

2.3.

Decomposition into

odd and even components,

(a)

Original

Even

From

part,

Eq. 2.4

(c)

function,

(6)

Odd part.

we observe

FIG. 2.4. Decomposition into

even and odd components,
(a) Unit step function.
(6)
Even part of unit step, (c)
Odd part of unit step.

that

s(-t)

= st(-t) + * (-0
- sXO - *o(0

(2.5)

Consequently, the odd and even parts of the signal can be expressed as

J#)- MK0 + *(-')]

*(o = two - K-m

(2 .6;

Consider the signal s(t), shown in Fig. 2.5a. The function s(-t) is equal
axis and is given in Fig. 2.5ft. We then
to s(t) reflected about the t

obtain st(t) and j(0 as

shown

in Figs. 2.5c

and

d, respectively.

Signals

and waveforms

13

rf-tf

-1l

(a)

*Jt)

lift)

-]I

-i
(d)

(e)

FIG. US. Decomposition into odd and even components from

s(t)

and s(t).

Now let us turn our attention to the continuity property of signals.

Consider the signal shown in Fig. 2.6. At t
T, the signal is discontinuous.

The

height of the discontinuity

is

f(T+) -f(T-

/(T+)

where

=A

(2.7)

= lim/(T+c)
-0

f(T-)
and

is

(2.8)

- lim/(T- )

a real positive quantity. In particular, we are concerned with

neighborhood of f = 0. From Eq. 2.8, the points

discontinuities in the

{
T
FIG. 24. Signal with discontinuity.

Network

24

analysis

and synthesis

Tl

FIG.

/(0+) and/(0-)

2.7. Signal

with two discontinuities.

are

/(0+)

- lim/(e)
-0

/(0-)

(2.9)

= lim/(-e)

For example, the square pulse in Fig. 2.7 has two discontinuities, at
Tt The height of the discontinuity at Tx is

Tx and

s(Tx +)

- s(T -) = K

Similarly, the height of the discontinuity at

2.1

(2.10)

Ta is K.

GENERAL DESCRIPTIONS OF SIGNALS

we consider various time domain descriptions of signals.

In particular, we examine the meanings of the following terms: time
constant, rms value, d-c value, duty cycle, and crest factor. The term, time
In this section

constant, refers only to exponential waveforms;

the remaining four

terms describe only periodic waveforms.

Time constant
important to know how quickly a
of
the decay of an exponential is the
useful measure
waveform decays.
waveform described by
exponential
the
time constant T. Consider

In

many

physical problems,

it is

r(t)

From a plot

Also

of r(t) in Fig.

2.8,

= Ke~*IT u(t)
we

see that

when

(2.11)

T,

r(T)

= 0.37KO)

(2.12)

r(4T)

= 0.02r(0)

(2.13)

Signals and

waveforms

25

1.00

075
r(t)

0.50

037

025

O02
5

FIG.

2.8.

Normalized curve for time constant

1.

Observe that the larger the time constant, the longer it requires for the
waveform to reach 37% of its peak value. In circuit analysis, common
time constants are the factors RC and RjL.

RMS Value
The rms or root mean square value of a periodic waveform e(t) is defined
as
Crms

i?f>*r

(2.14)

where T is the period. If the waveform is not periodic, the term rms does
not apply. As an example, let us calculate the rms voltage for the periodic

waveform in

Fig. 2.9.

Inns

4^r(T-)M>]r
(2.15)

-fair
= J2J3Av
D-C Value
The

d-c value of a

waveform has meaning only when the waveform

waveform over one period

is

periodic. It is the average value of the

oc

= -1

fT

TJ

e{t)dt
p

(2.16)

26

Network

analysis

and synthesis

2T

-A

FIG.

The square wave

2.9. Periodic

waveform.

in Fig. 2.1 has zero d-c value, whereas the

waveform in

Fig. 2.9 has a d-c value of

1

[AT

ATI

(2.17)

Duty cycle
The term duty cycle, D, is defined as the ratio of the time duration of the
positive cycle t9 of a periodic waveform to the period, T, that is,
(2.18)

T
becomes important in dealing with waveforms of the type shown in Fig. 2.10, where most of the energy is conThe rms voltage of the waveform
centrated in a narrow pulse of width f

The duty

cycle of a pulse train

in Fig. 2.10

is

,h

*--(iM'
(2.19)

= AJ7JT

An

(o

FIG. 2.10. Periodic waveform with small duty

cycle.

Signals and

waveforms

27

TT
FIG.

2.1 1. Periodic

waveform with zero d-c and small duty

cycle.

Wc see that the smaller the duty cycle, the smaller the rms voltage. The
square wave in Fig. 2.1 has a 50% duty cycle.
Crest factor
Crest factor1

is defined as the ratio of the peak voltage of a periodic

waveform to the rms value (with the d-c component removed). Explicitly,
for any waveform with zero d-c such as the one shown in Fig. 2.1 1
crest
factor, CF, is denned as

CF = -^-

-^-

or

rms

whichever
voltage

is

is greater.

(2.20)

rms

For the waveform

in Fig. 2.11, the peak-to-peak

defined as

e, p

(2.21)

<?

Since the waveform has zero d-c value

eJ*

= et(T-t )

(2.22)
(2.23)

Also,

eb

= eD

and

ea

= ePP(l -

The rms value of the waveform

emg

D)

(2.24)

is

= ( e( l ~ D? +
f

enWr-

t )\

*
(2.25)

G.

Justice,

'The

(Jan., 1964), 4-5.

eppy/D(l

D)

Significance of Crest Factor," Hewlett-Packard Journal, 15,

No. 5

Network

28

analysis and synthesis

Since crest factor

CF =

ejerm*,

we have

CF =

e(\

D)

ep jD(l -

D)

(2.26)

-Vi/J>-i
For example,

if

D=

CF
(2.27)

= ^100 -

UD =

10

10,000'

CF=VlO,000- 1~100

(2.28)

A voltmeter with high crest factor is able to read accurately rms values of
whose waveforms differ from sinusoids, in particular, signals with
low duty factor. Note that the smallest value of crest factor occurs for
the maximum value of D, that is, X>mx = 0.5,

signals

CFnto

= Vl/iW =1

(2.29)

THE STEP FUNCTION AND ASSOCIATED WAVEFORMS

Z3

The

unit step function w(f)

shown

(/)

=
=

in Fig. 2.12

is

defined as

<
f0

(2.30)

physical analogy of a unit step excitation corresponds to a switch S,

and connects a d-c battery of 1 volt to a given circuit,
t
the
as shown in Fig. 2.13. Note that the unit step is zero whenever

The

which closes at

u(t)

^o
Switch

lv

Network

T.
FIG. 2.12. Unit step function.

FIG. 2.13. Network analog of unit

step.

Signals

and waveforms

29

m
*<t-a)

a
FIG. 2.14. Shifted step function.

argument
argument

(/)

greater than zero.

> 0, is defined by
vit

and

is

occurs

FIG. 2.15. Square pulse.

within the parentheses

(t) is

- a) =
=

is

and is unity when the

the function u(t
a), where

negative,

Thus

/
1

<a

{Zil)

t^a

shown in Fig. 2.14. Note that the jump discontinuity of the step
when the argument within the parentheses is zero. This forms the

basis of the shifting property of the step function. Also, the height of the

jump discontinuity of the step can be scaled up or down by the multiplication of a constant K.

With the use of the change of amplitude and the shifting properties of
the step function, we can proceed to construct a family of pulse waveforms.
For example, the square pulse in Fig. 2.15 can be constructed by the sum
of two step functions
sif)

= 4u(t -

1)

The "staircase"
by the equation

as given in Fig. 2.16.

characterized

<

s(t)

=>J,u(t

(-4)

u(t

- 2)

function,

shown

(2.32)

in Fig. 2.17,

- kT)

*t)

-4
FIG.

2.16. Construction

of square pulse by step function.

is

(2.33)

Network

30

analysis

and synthesis

3
2

1
1

ZT

FIG.

2.17. Staircase function.

wave in Fig. 2.1. Using the

that the square wave is given by (for t ^ 0)

shifting

Finally, let us construct the square

property,
s(t)

we

see

m(0

simpler

way

2u(t

T)

2u(t

2T)

to represent the square

wave

2k(*
is

- 3D +

'

(2.34)

by using the property

zero whenever its argument is negative. Restricting

ourselves to the interval f ;> 0, the function
that the step function

is

s(0

is

by the waveform in Fig.

2. 1 can be represented
Fig.
in
wave
It is now apparent that the square

zero whenever sin (irtIT)

2.18.

(2.35)

-(*?)

S (0

is

negative, as seen

= (sin^)-u(-sin^)

(2.36)

Another method of describing the square wave is to consider a generalization of the step function known as the sgn function (pronounced signum).

The sgn function

is

defined as

sgn [/(f)]

=1
=

>o
fit) < o
fit)

fit)

s(t)

T
FIG.

2. IS.

2T
The

3T

AT

signal u(sin7rr/r).

(2.37)

Signals

and waveforms

31

ft)

ZT

3T

FIG. 2.19. Sine pulse.

Thus the square wave in

Fig. 2.1

s(0

is

simply expressed as

sgn

sin

(2.38)

Returning to the shifting property of the step function,

can be represented as

we

see that the

single sine pulse in Fig. 2.19

s(r)

- sin ^ [u(t - IT) - u(t -

3T)]

(2.39)

The

step function is also extremely useful in representing the shifted or

delayed version of any given signal. For example, consider the unit ramp
p(t)

tu(t)

(2.40)

shown in Fig. 2.20. Suppose the ramp is delayed by an amount t = a, as

shown in Fig. 2.21. How do we represent the delayed version of ramp?

FIG.2JM.

Ramp function with zero time shift.

Network

32

analysis

and synthesis

s(t)

Ramp function with

FIG. 2.21.

First, let

us replace the variable

When

p(t') is plotted against

of 5(0 versus

we

in Fig. 2.20.

t',

If,

When we plot p(t') against

t,

= a.
t'

f u(t')

a.

Then
(2.41)

the resulting curve is identical to the plot

t ' in p(t'),
however, we substitute t a

then have
p(f)

shift

by a new variable

p(r')

time

(t

- a) u(t -

(2.42)

a)

we have the delayed version of pit) shown in

Fig. 2.21.

From the preceding discussion, it is clear that if any signal /(/) u(t)
delayed by a time T, the delayed or shifted signal is given by

f{t')=f{t-T)u{t-T)
For example,

let

is

(2.43)

us delay the function (sin irtfT) u(i) by a period T. Then

the delayed function

s(t'),

s(t')

shown

[sin

in Fig. 2.22,

- (t -

is

T)] u(t

FIG. 2.22. Shifted sine wave.

T)

(2.44)

Signals

and waveforms

33

m2-

FIG. 2.23. Triangular pulse.

As a final example, consider the waveform in Fig. 2.23, whose component parts are given in Fig. 2.24. For increasing t, the first nonzero
component is

the function 2{t

1) u(t

1),

which represents the

straight

At ? = 2, the rise of the straight line is to be

arrested, so we add to the first component a term equal to 2(t 2)u(t 2)
with a slope of 2. The sum is then a constant equal to 2. We then add a

line

of slope 2 at

2u(t

term

s(t)

1.

2) to bring the level

= 2{t -

1) u(t

1)

down
2(t

to zero. Thus,
2) u(t

2)

2u(t

2)

(2.45)

1A THE UNIT IMPULSE

The unit impulse, or deltafunction, is a mathematical anomaly. P. A. M.
first used it in his writings on quantum mechanics. 2 He defined the

Dirac

2(t-\)u(t-\)

FIG. 2.24. Decomposition of the triangular pulse in Fig. 2.23.

P.

A. M. Dirac, The Principles of Quantum Mechanics Oxford University

Press, 1930.

Network

34

anal/sis

delta function d\t)

and synthesis

by the equations

=
<J(0 =

6\t) dt

f:
most important property

Its

is

(2.46)

for

(2.47)

the sifting properly, expressed by

f(t)d(t)dt=f(0)

f
J

(2.48)

we will examine the unit impulse from a nonrigorous

Those who prefer a rigorous treatment should refer to
Appendix B for development of this discussion. The material in that
appendix is based on the theory of generalized functions originated by
G. Temple. 8 In Appendix B it is shown that the unit impulse is the
In this section

approach.

derivative of the unit step

r

At

first

..

o(0

glance this statement

is

if

/0 ao>l
(2.49)

,.

(0

doubtful. After

all,

the derivative of the

zero everywhere except at the jump discontinuity, and

it does
not even exist at that point! However, consider the function g t (i) in Fig.
2.25. It is clear that as e goes to zero, g ( (t) approaches a unit step, that is,

unit step

is

lim ge(t)

(0

(2.50)

-o

Taking the derivative of g t(t), we obtain ',(0. which

is

defined by the

equations
g'(0

as

shown in

>

e 4+1 .

-e
= 0;

t<0, t>
Now let take on a sequence of values e

such that
(
Consider the sequence of functions {g' c< (0} for decreasing
Fig. 2.26.

*V0

et (t)

FIG.

(2.51)

US.

Unit step when

0.

FIG.

1M.

Derivative of

g((t)

in

Fig. 2.25.

G. Temple, "The Theory of Generalized Functions," Proe. Royal Society, A, 228,

1955, 175-190.

Signals and

waveforms

35

h-\W>

8it

s s

FIG. 127. The sequence {gfi (t)}.

values of c { , as

shown

in Fig. 2.27.

The sequence has the following

property:
J't>0g' (t)dt
(i
*1<0

where

tx

and

/,

are arbitrary real numbers.

there corresponds a well-behaved function

g't t(t).

As

(2.52)

For every nonzero value of e,

(i.e., it does not "blow up")

approaches zero,

g'(o+)-

00
,-0

(2.53)

so that the limit of the sequence is not defined in the classical sense.
Another sequence of functions which obeys the property given in Eq.
2.52

is

We

the sequence {///)}

Fig- 2.28.
now define the unit impulse
of all sequences of functions which obey Eq. 2.52. In

6{t) as the class

particular,

we

define
r*t>o
J*i>o

t)A-lim

i<0

g' t((t)dt

e<-0/i<0

a
4hm

r* >o

UQdt

(2.54)

Network

36

analysis

and synthesis

FIG. 2.28. The sequence {/ ,(0>

It

should be stressed that this

stated previously,

From

is

is

not a rigorous definition (which, as

found in Appendix B) but merely a

the previous definition

we can

heuristic one.

think of the delta "function" as

having the additional properties,

<5(0)

<5(f)

=
=

oo
(2.55)

for

t^O

waveforms

Signals and

Continuing with
the impulse

is

this heuristic treatment,

say that the area "under"

unity, and, since the impulse is zero for

d(t)dt

Thus

we

6
6{i)
dt

the entire area of the impulse

/*o-

S(t) dt

we have
(2.56)

is

"concentrated" at

r+oo
(5(f)

dt

=
is

0.

Con-

zero,

s(t)

J0+

J-oo

# 0,

Jo

<x>

sequently, any integral that does not integrate through

as seen by

37

s(t)

= Au(t- a)

(2.57)

A,

The change of scale and time shift properties

discussed earlier also apply for the impulse
function.

The
s(t)

yields

derivative of a step function

= Au(t a)

(2.58)

an impulse function
B'(t)

s\t)

= A6(t- a)

(2.59)

which is shown in Fig. 2.29. Graphically, we

represent an impulse function by an arrowhead pointing upward, with the constant

A written
Note that A is the
A d(t - a).

multiplier

s'(t)

= Ad(t-a)

next to the arrowhead.

FIG. 2.29

area under the impulse

Consider the implications of Eqs. 2.58 and 2.59. From these equations
see that the derivative of the step at the jump discontinuity of height A

we

an impulse of area A at that same point t = T. Generalizing on this

argument, consider any function /(0 with a jump discontinuity at / = T.
Then the derivative, /'(0 must have an impulse at t = T. As an example,

yields

consider /(0 in Fig. 2.30.

At

T,f(t) has a discontinuity of height A,

T
FIG. 2.30. Function with discontinuity at

T.

38

Network

which

is

analysis

and synthesis

given as

A =/(r+) -f(T-)
Let us define

same shape

fx(t)

(2.60)

as being equal to f(t) for

<

as/(/), but without the discontinuity for

T,

and having the

>

T, that

is,

Ut)=f(t)-Au{t-T)
The

derivative

A0

is

then

A0 -AC) + ^
The following example
the function /(f)

(2.61)

illustrates this

<K*

point

T)

(2.62)

more

clearly.

In Fig. 2.31a,

is

fit)

= Au(t-a)-A u(t - b)

(2.63)

Its derivative is

fit)

= Ad(t - a) - Ad(t - b)

(2.64)

and is shown in Fig.

2.316. Since/(/) has

two discontinuities, at t = a and

The coefficient of

b, its derivative must have impulses at those points.

the impulse at = b is negative because
t

f(b+) -fQ>-)

- -A

(2.65)

m
A

8(0
4

(a)

r<*>

(a)
g'(t)

A'

2
2

~0]

'-A

(b)

FIG. 2.31. (a) Square pulse,

of squairepulse.

(b) Derivative

FIG .2.32.

(a) Signal,

(*)] derivative.

Signals and

waveforms

39

As a second example,

consider the function g(i) shown in Fig. 2.32a.

obtain #'(0 by inspection, and note that the discontinuity at /
1
produces the impulse in #'(0 of area

We

0+)-(l-) =

2,

(2.66)

as given in Fig. 2.32*.

Another interesting property of the impulse function is expressed by the

integral

7(0 S(t-T)dt= f(T)

f
This integral
1

5* T.

is easily

evaluated if

that

consider that d(t

T)

for all

Therefore, the product

f{t)d(t-T)
If/(r)

we

(2.67)

is

we

single-valued at

aXLtitT

T,f(T) can be factored from the

(2.68)

integral so

obtain

/(T)

d(t

J O0

-T)dt= f(T)

Figure 2.33 shows /(r) and d(t - T), where /(/)

If/(0 has a discontinuity at / = T, the integral

is

(2.69)

continuous at i

T.

+00

Jf(t)Ht-T)dt

not denned because the value of f(T)

the following examples.
is

Example

is

not uniquely given. Consider

2.1

/(0

- 1**

T)dt

- e* r

+0O

Example

Z2

e
~&*Hf

(2.70)

-00

/(/)

= sin t

>''('- 5)*-^

(2.71)

Network

40

analysis

and synthesis

f(t)

FIG. 2.34. Impulse scanning.

Consider next the case where/ (0 is continuous for

us direct our attention to the integral

f(t)d(t-T)dt=f(T)

oo

< <
t

Let

oo.

(2.72)

-oo to +oo,
which holds for all / in this case. If T were varied from
of this sort
operation
An
entirety.
its
in
reproduced
then/(0 would be
paper with
of
sheet
a
moving
by
function/(f)
the
scanning
corresponds to
across a plot of the function, as shown in Fig. 2.34.
higher order derivatives of the unit step function.
Let us
we represent the unit impulse by the function ft(i) in Fig. 2.28,

a thin

slit

now examine

Here

-* 0, becomes the unit impulse. The derivative of/ (0 is given

derivative of the
in Fig. 2.35. As e approaches zero,/' (0 approaches the
in Fig. 2.36.
seen
unit impulse d'(t), which consists of a pair of impulses as
to zero.
equal
is
doublet,
The area under d'(t), which is sometimes called a

which, as

Thus,
6'(t)

r
The

dt

other significant property of the doublet

I
1

/() d'(t

(2.73)

is

-T)dt= -f\T)

rt (t)

f2

(2

FIG. 2.35. Unit doubles as

<=

-* 0.

FIG. 2.36. The doublet

d'(t).

(2.74)

Signals and

waveforms

41

where f'(T) is the derivative of/(0 evaluated at t = T where, again, we

assume that/(/) is continuous. Equation 2.74 can be proved by integration
by parts. Thus,

*/

f(t)d'(t-T)dt=f(t)d(t-T)\

=
It

(t)d(t-T)dt
-00

/ oo

(2.75)

-f'(T).

can be shown in general that

f
r
oo

/(O

-T)dt = (-l)y '(r)

<B,
<5

(<

(2.76)

where (5< n and/ <"> denote nth derivatives. The higher order derivatives of
can be evaluated in similar fashion.
>

d(t)

Problems
2.1

Resolve the waveforms in the figure into odd and even components.

s(t)

s(t)

-v

+1

(a)

a>)

s(t)

PRO B. 2.1

Network

42

analysis

and synthesis

2.2 Write the equation for the

functions.

waveforms

in the figure using shifted step

fit)

2
1

A
2

(b)

f(t)

fit)

1.5
1

(d)

(c)

PROB.

2.2

in Prob. 2.2 and write the equations

impulse functions.
and/or
step
shifted
using
for the derivatives,
2.4 For the waveform /(r) given in the figure, plot carefully

23

Find the derivative of the waveforms

for a value of

>

f(r)dr

T.

f(t)

f(t)

2*Wl
EMt)
>*

-E6(t-T)

PROB. 2.4

PROB. 2.5

Signals and

2^ For

the waveform /(r),

must be so that

shown

waveforms

in the figure, determine

43

what value

P/(0<//-0

()

J 00
j"7<f)<fr-o

<*)

J0+
2.6

For the waveforms shown

time functions, that

is,

/",(/

in the figure, express in terms of elementary

- Q, d(t -

/,),

CU.1.1.
Sketch *u_

r
*
rL\
j * x
..
the ..
waveforms
for (6) and (c) neatly.

()/(/) (*)/'(/) (c) f

/(T) Af
J 00

JW

m
20

"*

At

(0

(ii)

PROS.
2.7

2.4

Prove that

(a)

<*'(*)

(6)

--<}'(--X)

-<(a;)

= x d'(x)

/oo

(c)

;*)

&

-o

J 00
2.8

The waveform /(0

in the figure is defined as

/W-^(/-e)*,
= 0,
Show that

as e

;!

o<;/

elsewhere

0,/(/) becomes a unit impulse.

J" 5

Network

44

analysis

and synthesis

Plot

2.9
(a)

%s

(ft)

2.10

sgn (cos 0;

t 2n

'

Evaluate the following integrals

(fl )

- TO u(t - T2) dt;

<*t

Ti <T^

J CO
/go

a,)

<5(a>

a> )

cos

o>f rfo>

J 00
OO

J
2.11

[<5(0-/4<5(r-7'1 )+2/l<5(f-7 2)]e-^"A

(c)

00

Evaluate the following integral

f " sin

2.12

(t

- l\ +

S'

('

"

f)

The response from an impulse sampler

r(t)

Plot

tU

KO for

<, t <.

sin

n s(t

*'<'

>]

given by the equation

is

K = 0,

-Kj\dt;

1, 2, 3,

IT.

If the step response of a linear, time-invariant

determine the impulse response h(f), and plot.

2.13

system is r s (0

2.14 For the system in Prob. 2.13 determine the response

excitation

due to a

2e

'

u(t),

staircase

j(0-i(/-*r)
*=o
Plot both excitation

and response

functions.

impulse response of a time-invariant system

determine the response due to an excitation
2.15

If the

e(t)

Plot

2S(t

1)

2d(t

e(t).

PROB. 2.16

2)

is

W) = ~* "(')

Signals and

2.16

The

unit step response of a linear system

<x(r)

(a)

Find the response

(b)

Make

(2e-

2t

waveforms

45

is

1) (/)

r(f) to the input /(f).

a reasonably accurate sketch of the response.
dimensions.

Show

all

pertinent

chapter

The frequency domain:

Fourier analysis

3.1

INTRODUCTION
One of

signals.

the most

If

common

classes of signals encountered are periodic

T is the period of the signal, then

s(t)

= s(tnT)

7i

= 0,l,2,...

In addition to being periodic, if s(t) has only a

tinuities in any finite period and if the integral

"T
I
is finite

(where a

is

an arbitrary

\s(t)\

(3.1)

number of discon-

finite

dt

can be expanded

real number), then s(t)

into the infinite trigonometric series

s(t)

=^+
+

+ a t cos 2<ot +
b x sin cot + b t sin "hot +

a x cos

tot

2\

m=

2w/r. This trigonometric series is generally referred to as the

Fourier series. In compact form, the Fourier series is

where

s(t)

= ?* + 2 ( cos not +
2

b n sin

ntot)

(3.3)

apparent from Eqs. 3.2 and 3.3 that, when s(t) is expanded in a
Fourier series, we can describe s(t) completely in terms of the coefficients

It is

46

The frequency domain: Fourier

analysis

47

harmonic terms, a* alt a . , bx, b

These coefficients cona frequency domain description of the signal. Our task now is to
derive the equations for the coefficients a( , b t in terms of the given signal
function s(t). Let us first discuss the mathematical basis of Fourier series,
the theory of orthogonal sets.
of

its

stitute

ORTHOGONAL FUNCTIONS

3.2

Consider any two functions

Then

and/(/) that are not identically zero.

x (f)

if

r,

f
JTx

/i(0/*(0dt

=O

(3.4)

we

say that/i(r) and/^/) are orthogonal over the interval [Tlt TJ. For
example, the functions sin t and cos t are orthogonal over the interval
nlit

r <;

<f>&)

(
<j>

H(t)}.

Consider next a set of real functions {&(*),

l)2n.

If the functions obey the condition

(*.

4>i)

V<(0

4>M dt =

i*j

0,

(3.5)

then the set {<f>t } forms an orthogonal set over the interval [7\, TJ. In
Eq. 3.5 the integral is denoted by the inner product (^ 4 , <f> ). For convent
ience here, we use the inner product notation in our discussions.
The set {^J is orihonormal over [Tu 7",] if

=
The norm of an element

l&l

<f>

k in

the set

3.1. The Laguerre

time domain approximation,

W. H.

t-.j

{^J

is

defined as

- (4, &)* - (%\0 *)*

We can normalize any orthogonal

term <f>k by its norm ||^J|.
'

(3.6)
1

set

{<f>x, <f>t,

. . ,

fa} by dividing each

set,1
is

which has been shown to be very useful in

orthogonal over [0, <]. The first four terms

Kautz, "Transient Synthesis in the Time Domain," Trans.

No. 3 (Sept. 1954), 29-39.

Theory, CT-1,

(3-7)

IRE on

Circuit

Network

48

analysis

and synthesis

of the Laguerre set are

- e~**

*i(0

Ut) -

To show

- 2(fl01

*,(/)

= e-o'H - MfO) + Hatf\

4>t(t)

= e-'[l -

far

(3.8)

+ 6(flf)* - Ka/)8

that the set is orthogonal, let us consider the integral

f"

Letting t

e-*[l

Ut) hiO * = f V*"[l - MpQ +

2(0*] dt

(3.9)

= at, we have

i,-;;J'V a-4T+2,*>rfr
,,

(3.10)

-ill-"]The norms of &(/) and &(/) are

,_(J[-^.*)
-

11*1

-^

"^[l ~ 4(0 + 4<a

a
') ]

Uo

(3,.,

*}
J

(3.12)

V2a
It is

not

difficult

to verify that the

norms of all the elements

equal to l/V2aT Therefore, to render the Laguerre

each element fa by

3.3

set

in the set are also

orthonormal,

we

divide

1/ V2a.

APPROXIMATION USING ORTHOGONAL FUNCTIONS

In this section

we

explore

some of the uses of orthogonal functions in

The principal problem is that of

the linear approximation of functions.

approximating a function/(0 by a sequence of functions/B(f) such that the

mean squared error

e

= lim
n-oo

*W) - /(0]

f
JTi

dt

(3.13)

The frequency domain: Fourier

When

Eq. 3.13

we

is satisfied,

49

analysis

say that {/(f)} converges in the mean to

fit).

To examine

the concept of convergence in the

mean more

must first consider the following definitions:

Definition 3.1 Given a function/(f) and constant/;

>

we

closely,

for which

T,

\f(t)\*dt<ao,

(
JTi

we

say that /(f)

[T

Tt

is

integrable

L*

in [7\,

TJ, and

we

write /(/)

e L

in

].

Definition

12

tions integrable

If/(f)

eL'in [Tlt Tt], and {/(f)> is a sequence of func-

L* in [Tlt TJ, we say

that if

Bm.P|/(0-/^)r*-0
then

{/.(*)}

The

mean of order/? to /(f). Specifically, when

mean to /(f).

converges in the

2 we say that

{/(f)} converges in the

principle of least squares

Now let us consider the case when/(f) consists of a linear combination

of orthonormal functions

4>i,<l>

/(0

<f>

n-

= I *<&(<)

(3.14)

<-i

Our problem

is

to determine the constants at such that the integral

squared error
11/

- All* - f *W) - M9f dt

(3.15)

JTx

is

a minimu m. The principle of least squares states that in order to attain

( must have the values

m inimum squared error, the constants a

Ci

Proof.

must

set

= ]Kt)Mt)dt

We shall show that in order for

a{

ct for every

H/-/.II*

1, 2,

. .

,
.

\\f

(3.16)

/|| s to be minimum

we

n.

= </,/) - 2{f,fJ + (/,/)

=

11/11*

-22a,(/,&)+2a,W

(3.17)

Network

50

Since the set

(/,&).

analysis

{&}

and synthesis

orthonormal, |\| a

is

1,

and by

definition, ct

We thus have
I/-/-I'

Adding and subtracting

i/-/.!"

l/l'

- 22A + i

(318)

'*

2 c* gives
8

ii/n

- i*"
- 2iA
+ i*'
+ i^
_1
<-l
<-l
<_1

(3.19)

We see that in order to

at

set

= c<.

coefficients

The

minimum

attain

integral squared error,

we must

defined in Eq. 3.16, are called the Fourier

coefficients c t ,

of/(*) with respect to the orthonormal

Parseval's equality
Consider fn(t) given in Eq. 3.14.

set

{+&)}.

We see that

ft

I L/j(or*-i'
since

320>

This result is known as Parseval's

important in determining the energy of a periodic signal.

& are orthonormal functions.

equality,

3.4

<

and

is

FOURIER SERIES

Let us return to the Fourier series as denned earlier in this chapter,

aft)

_ s + V (a , cos nmt +
2

b n sin nmt)

(3.21)

Ti

From our discussion of approximation by orthonormal functions, we can

by a
see that the periodic function s(t) with period T can be approximated
is,
that
s(t),
to
mean
the
in
Fourier series sn(t) such that s n(t) converges
lim
im
-oo

where a

is

any

real

mean squared error

r^MO-s-O)]*^

number.
||j(0

322>

Ja

We know, moreover, that if n

- *(0II* is minimized when

is finite,

are the Fourier coefficients of s(t) with respect to the orthonormal set
/cos koat

sin k<ot \

.__,

the

the constants a{ , b t

The frequency domain: Fourier

51

analysis

fYVtW\

-3x

-2x

-x

FIG.

2x

3.1. Rectified sine

3x

wave.

in explicit form the Fourier coefficients, according to the definition given

obtained from the equations

earlier, are

*)A

fl0=s

rJ
2 r

s(t)

-If
mean, when

s(t)

cos kmt dt

s(0 sin tor

We should note that because the

s(t) in the

(323)

/f

(3.24)

(3.2S)

Fourier series sn(t) only converges to

jump discontinuity, for example,

contains a

= ^IJ(o=)

S .0,)

( 3. 26)

At any point

tt that s(t) is differentiable (thus naturally continuous)

converges to a(/J.*
As an example, let us determine the Fourier coefficients of the fully
rectified sine wave in Fig. 3.1. As we observe, the period is
n so
that the fundamental frequency is m
2. The signal is given as

*(*i)

T=

s{t)

Let us take a
derived,

= A |sin

and evaluate between

we have

(3.27)

t\

and

ir.

Using the formula just

K = ~ Jo (0 sin 2nt dt =
|

(3.28)

TT

2A C*
&A
=
sin tdt =
\

IT

Jo

2[>

(3.29)

s(0 cos 2nt dt

IT

Jo

-An*

AA
(3.30)

For a proof see H. F. Davis Fourier Series and Orthogonal Functions,

Allyn and
Bacon, Boston, 1963, pp. 92-95.

Network

52

analysis

Thus the Fourier

of the

series

5(/)

and synthesis

wave

rectified sine

- i(l + f

is

cos 2nt)

(3.31)

EVALUATION OF FOURIER COEFFICIENTS

3.5

In this section we will consider two other useful forms of Fourier series.
In addition, we will discuss a number of methods to simplify the evaluation
of Fourier coefficients. First, let us examine how the evaluation of

symmetry considerations. From Eqs. 3.23-3.25

which give the general formulas for the Fourier coefficients, let us take
a = T/2 and represent the integrals as the sum of two separate parts,

coefficients is simplified by

that

is,

rr/t

T
ss

ro

s(0 cos nmt dt

s(t)

cos mot dt

J
"

J-T/S
T/a

JO

2 rf

-i

T,t

nmt dt

s(t) sin

TUo

s(f) sin

(3.32)

nmt dt

J-T/i

Since the variable (0 in the above integrals is a dummy variable, let us

t in the
substitute x
t in the integrals with limits (0 ; r/2), and let x

integrals with limits

an

bn

(T/2;

= 2

T/ *
[s(x)

TJ

T Jo

+ s(x)] cos nmx dx

(3.33)

T/
2 C
I

Then we have

0).

*
[s(a:)

s( *)] sin nmx dx

Suppose now the function is odd, that is,

a =

for all n,

s(x)

s(x), then we

see that

and
T/2

-if*

nmx dx

s{x) sin

tJo

This implies that,

sine terms.
s(x)

On

if a

function is odd,

its

Fourier series will contain only

the other hand, suppose the function

= s(x), then b n =
an

(3.34)

is

even, that

is,

and

TJo

s(x) cos

nmx dx

(3.35)

Consequently, the Fourier series of an even function will contain only

cosine terms.

The frequency domain: Fourier

anal/sis

53

FIG. 3.2

Suppose next, the function

s(t)

,(,

obeys the condition

l)--<0

by the example in Fig. 3.2.

only odd harmonic terms, that is,

as given

and

Then we can show

= bn = 0;

-if
TJo
T Jo

s(t)

(3.36)

that s(t) contains

neven

cos nmt dt
(3.37)

5(0 sin nmt

n odd

dt,

With this knowledge of symmetry conditions, let us examine how we

can approximate an arbitrary time function s(t) by a Fourier series within
an interval [0, 71. Outside this interval, the Fourier series sn(t) is not
required to fit s(t). Consider the signal s(t) in Fig. 3.3. We can approximate s(t) by any of the periodic functions shown in Fig. 3.4. Observe
that each periodic waveform exhibits some sort of symmetry.
Now let us consider two other useful forms of Fourier series. The first
is the Fourier cosine series, which is based upon the trigonometric identity,

C cos (na>i + 8J

C cos mot cos 0

^|

C sin nmt sin 0

FIG. 3J. Signal, to be approximated.

(3.38)

Network

54

anal/sis

and synthesis

m
A
^
^

^*^T

-T--

^"'

(a)

m
A
"*-..

-r'- "~^

f~"-~.

-A
(6)

m
A
**-

-T)

t
*

I**"

-A
(e)

FIG. 3.4. (a) Even function cosine terms only, (ft) Odd function
Odd harmonics only with both sine and cosine terms.

sine terms only,

(c)

We can derive the form of the Fourier cosine series by setting

= C cos 0
bn = C sin 0
a

and

We then obtain Cn and 0 in terms

0,

(3.40)

of an and bn, as

c,-(fl.'

Ca =

(3.39)

a
(3.41)

(-!:)

combine the cosine and sine terms of each harmonic in the original
we readily obtain from Eqs. 3.38-3.41 the Fourier cosine series

If we
series,

f(t)

= C + Ct cos (mi + 00 +
+ C.cos (3fl>f + 0,) +

C, cos (2o>f

0.)

+ Cn cos(nmt + J +

(3.42)

The frequency domain: Fourier

55

anal/sis

should be noted that the coefficients C are usually taken to be positive.

a term such as 3 cos 2cot carries a negative sign, then we can
use the equivalent form
It

If however,

3cos2a>f =
For example, the Fourier
was shown to be

of the fully

series

= ^(l + f

s(0

2ir\

Expressed as a Fourier cosine

s(0

3 cos (2<u/

n=i

+ tt)

(3.43)

rectified sine

wave in

Fig. 3.1

-^
cos 2nt)
An

(3.44)

series, s(t) is

= ^[l + 1 cos (2nt + *)]

2wL
n-i4n 1

(3.45)

Next we consider the complex form of a Fourier series. If we express

cos ncot and sin mot in terms of complex exponentials, then the Fourier
series

can be written as

(3.46)

= + 2 ( "* ~ J&" **"* +

2

fl

" + ^* *-'"')

n-l\
n-l

If we define

"~ n

'

'

then the complex form of the Fourier series

is

= A + i(/ffe*"" + /*--'"")

5(0

"^

- 2
n

34)

PSoo

We can readily express the coefficient fi as a function of s(t), since

/?

Pn

= l

~ Jb
C

T Jo

=i

f
T Jo

Equation 3.49

and Eq. 3.48

is

is

s(0(cos nott

s(t)e-

j sin na0 dt

(3.49)

mt dt

i'

sometimes called the discrete Fourier transform of

the inverse transform of P(na))

/?.

(/)

56

Network

analysis

and synthesis

Cb

Ci

Ci

C2

Ch

<h

Ci\

c*.

-Au -3 -2w -w

FIG.

3.5.

2(i>

Amplitude spectrum.

-2T

2T

HF -T 4
-A

FIG.

3d)

3.6.

Square wave.

,C*
4a>

The frequency domain: Fourier

Observe that /?

is

usually complex

/*n

The

real part

of fi

Re /?,

- -1

and the imaginary part of /?

T Jo

+yIm,

(3.50)

(0 cos tuot dt

(3.51)

is

T Jo

It is clear that

and can be represented as

CT

JImfl,-^f
function in n.

57

obtained from Eq. 3.49 as

is

Re Pn

Re/8B

anal/sis

s(0sinn>r<fc

(3.52)

Re /? is an even function in r, whereas Im /S is an odd

The amplitude spectrum of the Fourier series is defined as

ImPM

|/JJ-(Re"fc.+
and the />/uue spectrum

is

(3.53)

denned as

6,

= arctan^

(3.54)

It is easily seen that the amplitude spectrum is an even function and the
phase spectrum is an odd function in n. The amplitude spectrum provides
us with valuable insight as to where to truncate the infinite series and still
maintain a good approximation to the original waveform. From a plot of
the amplitude spectrum, we can almost pick out by inspection the nontrivial terms in the series. For the amplitude spectrum in Fig. 3.5, we see
that a good approximation can be obtained if we disregard any harmonic
above the third.
As an example, let us obtain the complex Fourier coefficients for the
square wave in Fig. 3.6. Let us also find the amplitude and phase spectra
of the square wave. From Fig. 3.6, we note that s(t) is an odd function.
Moreover, since s(t
T/2) = s(t), the series has only odd harmonics.
From Eq. 3.49 we obtain the coefficients of the complex Fourier series as

Ae-'Mi

--

TJo

dt--\

Ae' ,nat dt

TJt/
(3.55)
(I

2e~ linmT/i)

-)-

jnmT
Since ncoT

e~

,tuaT
)

= nlir, /? can be simplified to

/.

- rr- a - 2*"*" +

"*"*)

( 3 - 56 >

58

Network analysis and

synthesis
Amplitude

-3

-5

<a)

Phase

-5

-3

-1
2

(b)

FIG.

3.7. Discrete spectra

Simplifying

/5

one step

of square wave, (a) Amplitude. (*) Phase.

we

further,

obtain

n odd
(3.57)

jnir

=
The amplitude and phase

n even

spectra of the square

wave are given in

Fig. 3.7.

OF FOURIER COEFFICIENTS USING

UNIT IMPULSES

34 EVALUATION

In this section

we make use of a basic property of impulse functions to

is
simplify the calculation of complex Fourier coefficients. This method
restricted to functions which are made up of straight-line components only.
Thus the method applies for the square wave in Fig. 3.6. The method is

based on the relation

f" /(*)(-

J 00

TO* -/(TO

(3.58)

Let us use this equation to evaluate the complex Fourier coefficients for
,fmt
e~
, we have
the impulse train in Fig. 3.8. Using Eq. 3.58 with/(f)

'

~ *~ 2V,,

e
d%
T fH' - 2 V'"" '~T

<

(359>

The frequency domain: Fourier

Al

A..

A,i

_L

1
2

FIG.

3.8.

59

analysis

_L
3r
T

2T

Impulse

sr

train.

We

see that the complex Fourier coefficients for impulse functions are
obtained by simply substituting the time at which the impulses occur into
the expression, e~,m" t .
In the evaluation of Fourier coefficients, we must remember that the

P integral are taken over one period only, i.e., we consider

only a single period of the signal in the analysis. Consider, as an example,
the square wave in Fig. 3.6. To evaluate /?, we consider only a single
period of the square wave, say, from t
to
limits for the

T, as

wave

is

shown in Fig. 3.9a. Since the square

not made up of impulses, let us

differentiate the single period

wave to
can

give

s'(t),

now evaluate

as

shown

of the square

in Fig. 3.96.

We

the complex Fourier coeffi-

cients for the derivative s(t),

made up of impulses
s(t) is

>(t)

which clearly

is

alone. Analytically, if

-A

given as

s(0

(<)

- 1

inert

Pe

(3.60)

00

then the derivative of 5(0

W
is

A|

(3.61)

Here,

we

define a

new complex

coefficient

(3.62)

or

jnm

(3.63)

-2A
a function which
consists of impulse components alone, then
If the. derivative s'(i)

we

is

simply evaluate y n first and then obtain

3.63. For example, the derivative
of the square wave yields the impulse train

/?

from Eq.

(6)

FIG. 3.9. (a) Square wave

over period [0, 71. (b) Derivative of square wave over
period

[0, T\.

Network

60

and synthesis

analysis

in Fig. 3.96. In the interval

s'(t)

Then the complex

[0,

A6(t)-

T], the signal s'(f ) is given as

d(t

2^4

- |) + A d(t -

T)

(3.64)

coefficients are

Tvo
(3.65)

= (\ 2e- linmT/* +
}

e~

inmT
)

The Fourier

of the square wave are

coefficients

y
jnco
(3.66)

(i

2e~

linaTI * )

e~

inmT
)

jnmT
which checks with the solution obtained in the standard way in Eq. 3.55.
If the first derivative, /(f), does not contain impulses, then we must
differentiate again to yield

= f
n

s (f)

(3.67)

(3 - 68)

For the
[0,

triangular pulse in Fig. 3.10, the second derivative over the

T]

is

s(0

The

inmt

K =jno>Yn = (P )* P

where

period

Xne
oo

coefficients

= ^Uf) - 2d(f - |) +

A n are

Xn

now

=-

**

- d]

( 3 - 69 >

obtained as

T
( s"(i)e-

inat

dt

TJo
(3.70)

2A
which

,j

_ 2e-OfK27*>

-fnmT\

simplifies to give

X =s
"

&A

T*

.,

n odd
(3.71)

n even

The frequency domain: Fourier

analysis

61

A&1

3T
T

(a)

24 Y0

2"

2A

T
2A

(b)
[>"<'>

2A

(C)

FIG. 3.10. The triangular wave and

From

A_

we

its

derivatives.

obtain

P-r.

(Jmnf

2A

(3.72)

nodd

nV

neven

slight difficulty arises if the expression for s'(t) contains an impulse

in addition to other straight-line terms. Because of these straight-line

terms

we must

differentiation,

differentiate

once more.

However, from

we

this additional

obtain the derivative of the impulse as well.

presents no difficulty, however, because we know that
%(r) d'(t

T)

J GO

-s\T)

(3.73)

so that

f " d'(t
J 00

T)e- im"dt

This

= jna>e-"u T
'

(3.74)


Network

62

analysis

and synthesis

We can therefore tolerate doublets or even higher derivatives of impulses

la. Its derivative
in the analysis. Consider the signal s(f) given in Fig. 3.1
as
expressed
be
*'(<), shown in Fig. 3.116, can

S '(0

The second

M (0

|[

u(t

f )]

derivative s"(t) consists of

+ 0 - 2(l - f)

0.75)

a pair of impulses and a pair of

doublets as given by

mas

shown

|[aco

in Fig. 3.11c.

*(t

f )]

+ *W -

2d'(t

1)

We therefore evaluate A as
,-inmt

s"(t)e

dt

tJo

(3.77)

2. (i _ e-(*~r/) + 1 (i _ 2e- UnaTin )

s'(t)

.,

'

(3.76)

(W

(o)

-*'('-!)
"W
pit)

?(*-i)

to
FIG. 3.11

t-M(l-f)

The frequency domain: Fourier

The complex

coefficients /S are

now

63

analysis

obtained as

Qtonf

(i _ e-oZ7) + _!_ n _ 2e-(^r/a))

janT

(jconT)*

Simplifying,

(3-78)

'

we have
1

Pn=-
mr +
t

"odd

j2irn
(3.79)

n even
j2irn

In conclusion,

it

must be pointed out that the method of using impulses

to evaluate Fourier coefficients does not give the d-c coefficient, a /2 or

/3

We obtain this coefficient through standard methods as given by Eq. 3.23.

3.7

THE FOURIER INTEGRAL

In this section

we

extend our analysis of signals to the aperiodic case.

We show through a plausibility argument that generally, aperiodic signals

have continuous amplitude and phase spectra. In our discussion of Fourier
series, the complex coefficient (i n for periodic signals was also called the
discrete Fourier transform
1

CT
s(i)e-

M"'*dt

TJ-i
-T/2
and the inverse

(discrete)

transform was

*(0

From

= 2
n=

ln,'

Knfo)e

(3.81)

oo

the discrete Fourier transform

spectra which consist of discrete lines.

in the spectrum

(3.80)

we obtain amplitude and phase

The spacing between adjacent lines

is

A/ -

(n

l)/

=i

(3.82)

As the period T becomes larger, the spacing between the harmonic lines in
the spectrum becomes smaller. For aperiodic signals, we let T approach
infinity so that, in the limit, the discrete spectrum becomes continuous.

64

Network

and synthesis

anal/sis

We now define the Fourier integral or transform as

s(/)

lim
Tx>

W = r
Jo

A/-0

The

inverse transform

-"

s(()c

it

(3.83)

J oo

is

<0-f"
J

S(J)e

iU,t

(3.84)

df

00

Equations 3.83 and 3.84 are sometimes called the Fourier transform pair.
5~x denote
denote the operation of Fourier transformation and
If we let
inverse transformation, then

*</>-*-<0
s(t) = ?-i-S(f)
In general, the Fourier transform S(f)

S(f)

The

real part of

S(f)

is

is

(3 . 85)

complex and can be denoted as

= ReS(f)+ j Im S(f)

(3.86)

obtained through the formula

ReS(/)

J[S(/)

|"

cos 2-nft dt

S(-/)]
(3.87)

s(t)

J CO

and

the imaginary part through

= [SCO -

Im S(f)

S(-/)]

2/

(3.88)

=_

J CO

The amplitude spectrum of S(f)

A(f)

and the phase spectrum

is

s (0 sin 2tt/(

dt

defined as

[Re SC/) 2

+ Im (/)*]*

(3.89)

is

"-" Sin

**

Using the amplitude and phase definition of the Fourier transform, the
inverse transform can be expressed as
s(t)

A(f) cos

J 00
Let us examine some examples.

[lirft

<Kf)] df

(3.91)

The frequency domain: Fourier

MS)

65

anal/sis

<Kf)

-f

+/

(a)

((,)

FIG. 3.12. Amplitude and phase spectrum of A d(t

Example

/,).

3.2.*
sit)

= A6\t-td

S(f)~ ("

A^t- t9)er^t* dt

J 00

- Ae-i**">
Its

amplitude spectrum

(3.92)

is

A(f)=A
while

its

phase spectrum

(3.93)

is

<Kf)

(3.94)

-2^//o

as shown in Fig. 3.12.

Example 33.
Formally,

Next consider the rectangular function plotted in Fig.

we define the function as

HSyW

\x\

>

inverse transform of rect/is defined as sine

-1

3r

[rect/]

(3.95)

-W

recta;/

The

3.13.

the rect function.

(pronounced

sink),

sine t
W/2

e>*"*df

(3.96)

-I-Jf/2

sinirWr
Vt
* It should be noted here that the Fourier transform of a generalized function is also
a generalized function. In other words, if e C, (^ <f>) C. For example, J" <5(f ) =
1, where 1 is described by a generalized function 1(/). We will not go into the formal
details of Fourier transforms of generalized functions here. For an excellent treatment
of the subject see M. J. Lighthill, Fourier Analysis and Generalized Functions, England,
Cambridge University Press, 19SS.

<f>

Network

66

analysis and synthesis

recti

1.0

-*

"

FIG. 3.13. Plot of rect function.

1.0

0.8
0.6
sine

0.4

0.2

\\

0.0

-0.2
-0.3

~w

3.

~w

1.
~w

FIG. 3.14. The sine

e(t)

W
/

2.

3.

4.

5.

curve.

r(t)

Excitation

System

Narrow pulse

Wide band

*(0
Excitation

i.

-L

~w

Narrow pulse

r(t)

System

Response
10 Ti

10 Tx
Wide pulse

Narrower band

Wide pulse

FIG. 3.15. Illustration of the reciprocity relationships between time duration and
bandwidth.

The frequency domain: Fourier

67

anal/sis

From the plot of sine t in Fig. 3.14 we see that sine t falls as does |r| _1 ,
with zeros at t
n\W, n
We also note that most of the
1, 2, 3, .
energy of the signal is concentrated between the points
\\W t
\\W.

< <

Let us define the time duration of a signal as that point, t , beyond which
the amplitude is never greater than a specified value, for example, e
We
can effectively regard the time duration of the sine function as t
1/ W.
The value W, as we see from Fig. 3.13, is the spectral bandwidth of the
.

rect function.

We

see that if

W increases,

decreases.

The preceding

example illustrates the reciprocal relationship between the time duration

of a signal and the spectral bandwidth of its Fourier transform. This
concept

narrow

is

quite fundamental.

pulses,

i.e.,

It illustrates

why

in pulse transmission,

those with small time durations, can only be trans-

mitted through filters with large bandwidths; whereas pulses with longer
time durations do not require such wide bandwidths, as illustrated in
Fig. 3.15.

3.8

PROPERTIES OF FOURIER TRANSFORMS

In this section

we

some important

consider

properties of Fourier

transforms.

Linearity

The

linearity property

transform of a

of Fourier transforms states that the Fourier

signals is the sum of their individual Fourier

sum of two

transforms, that

is,

Hcx *i(0 + c

= Cl S (/) + c

t sjflj]

St(f)

(3.97)

Differentiation
This property states that the Fourier transform of the derivative of a
signal isjlnf times the Fourier transform of the signal itself:

J-s\i)=j2-nfS{f)

(3.98)

?- s (t) = (j2*frS(f)

(3.99)

or more generally,

The proof is obtained by taking the

derivative of both sides of the inverse

transform definition,
00

s'(0

= 7;f

S(J)e*<*df

at J -oo
(3.100)

-f
J

CO

j2nfS(f)e*"df

Network

68

Similarly,

it is

and synthesis

analysis

easily

that the transform of the integral of s(t)

shown
f

5(t)

'

dr\

= ^-

(3.101)

S(J)

j2nf

LJ-oo

is

Consider the following example

j(0
Its

Fourier transform

= -'(0

( 3 - 102>

is

s(/)

r^vHtyr^dt

f"CO
J

(3.103)

The

derivative of s(t)

s '(t)
Its

Fourier transform

+ j2irf

is

3(0

- aer** (0

(3.104)

is

?['')]

-1-

j2nf

a+j2wf

a+j2irf

(3.105)

= j2nfS(f)
Symmetry
The symmetry property of Fourier transforms

states that if

= X(f)
^ X(t) = *(-/)
3"

then

(3.106)

x(t)

(3.107)

This property follows directly from the symmetrical nature of the Fourier
transform pair in Eqs. 3.83 and 3.84.

Example

3.4.

From the preceding

^
It is

section,

we know

that

= rect/

(3.108)

= sine ( -/) = sine/

(3.109)

sine /

then simple to show that

&

rect

which conforms to the statement of the symmetry property.

the Fourier transform of the unit impulse,

property

&

8(t)

1.

From

Consider next
the symmetry

we can show that

8(f)

(3.110)

shown in Fig. 3.16. The foregoing example is also an extreme illustration

of the time-duration and bandwidth reciprocity relationship. It says that zero
time duration, 8(t), gives rise to infinite bandwidth in the frequency domain;
while zero bandwidth, 8(f) corresponds to infinite time duration.

as

The frequency domain: Fourier

analysis

69

HD

-/

-t

FIG. 3.1*. Fourier transform of /(f)

1.0.

Scale change

The

scale-change property describes the time-duration

-M*0

[.(*)]
Proof.

We prove

and bandwidth

It states that

reciprocity relationship.

this

(3.111)

property most easily through the inverse trans-

form
J-\\a\ S{af)]

\a\

f" S(af)e* df

(3.112)

J oo
Let/'

= af;

then

*^ll

S(f')]

\a\

f " S(/V*"'
J co

( '/o,

a
(3.113)

As an example,

consider

T[e* "(01

(3.114)

J2vf+a

then

Sy

\a\

(01
j2iraf

a
(3.115)

j2nf+\
ifo>0.
Folding

The

folding property states that

J-W-01

= s(-f)

(3.116)

Network

70

anal/sis

The proof follows

An

example

and synthesis

directly

from the

definition of the Fourier transform.

is

3=V u(-t)]

(3.117)

l-j2vf

Delay
is delayed by an amount t in the time domain, the corresponding effect in the frequency domain is to multiply the transform of the
undelayed signal by er iz"ft that is,

If a signal

<>,

(3.118)

For example,
&[e- aU

-' )

u(t-t

)]

-}2rftt>

=
a

(3.119)

+ j2nf

Modulation
The modulation or frequency

shift property of Fourier transforms

a Fourier transform is shifted in frequency by an amount f
the corresponding effect in time is described by multiplying the original

states that if

signal

by e iufot that
,

is,

3^[S(f - /<,)]

mM

(3.120)

s(t)

Example

3.5.
Given S(f) in Fig. 3.17a, let us find the inverse transform of
5"" 1
Si(f) in Fig. 3.176 in terms of s(t) =
S(f). We know that

Sxif)

S(f -f )

+ 5(/ + /)

(3.121)

-/

+/

-/

*f

(b)

FIG.' 3.17.

Demonstration of amplitude modulation.

The frequency domain: Fourier

Then

analysis

3^ 5,(/) - *>*". s{t) + er'*'U* g(t) = 2s(t) cos 2nfy

71
(3.122)

Thus we see that multiplying a signal by a cosine or sine wave in the time domain
corresponds to shifting its spectrum by an amount
f In transmission terminology / is the carrier frequency, and the process of multiplying s(t) by cos
.

2itfy

is

called amplitude modulation.

Parseval's

theorem

An

important theorem which relates energy in the time and frequency

domains is Parseval's theorem, which states that

"
i(0 St(0 dt
I
J
00

= f"

The proof is obtained very simply

f"

si(0 i(0 dt

J no

Sx(/) S,(-/) df

as follows:

f " s(t) dt f "

-f"
*

(3.123)

SiC/V""* d/

Si(/)4rf"

s&Vu"dt

(3.124)

Jao

oo

sMSA-fldf
-f"
J
00

when ^(r) = *j(f), we have a corollary of Parseval's theorem

known as PlancheraVs theorem.
In particular,

f " aty)
J 00

* = J
f"

ISC/)!

4T

(3.125)

00

If j(/) is equal to the current through, or the voltage across a

resistor, the total

energy

1-ohm

is

(0 dt

We see from Eq. 3.125 that the total energy is also equal to the area under
the curve of \S(f)\*.
or energy spectrum.

Thus

\S(f)\* is

sometimes called an energy density

Problems
3.1 Show that the set {1, sin nnt/T, cos nnt/T}, n = 1,2, 3,
, forms an
orthogonal set over an interval [a, a + 2T], where a is any real number. Find
the norms for the members of the set and normalize the set.
.

Network

72
3.2

analysis

and synthesis

Fourier
Given the functions/^) and/i(0 expressed in terms of complex

senes

/l(/)=2^
n oo

"

B,,

/K0-2A.'*-'
where both/i(0 and/j(0 have the same period T, and

show

= Kl e**\

A.

e*

that

oA>

+ 22 lgl cos(flB -

Note that
For the periodic

4>J

m * -n

\0,

33

lftl

signals in the figure, determine the Fourier coefficients

f(f>

--

+T

-T
(h)
fit)

T
5
(e)

PROB.3.3

A
T

The frequency domain: Fourier

analysis

3.4 For the waveforms in Prob. 3.3, find the discrete amplitude
spectra and plot.

73

and phase

3.5 For the waveforms in Prob. 3.3 determine the complex Fourier coefficients
using the impulse function method.
3.6

Find the complex Fourier

coefficients for the function

shown

figure.

A AA

ST,

3T.

~r

Tj

T,

-t

PROB.
3.7

L,

3T,

51

~T

~i

3.6

Find the Fourier transform for the functions shown in the

figure.

f(t)

/
/
-T

^C*

*T

(b)

(>

f(t)

f<t)

V
.1

3V

(c)

Id)

PROB.
3.8

3.7

Find the Fourier transform for

(a)

/ - A <3(0

(*)

/(f)

A sin atf

in the

74

Network

anal/sis

and synthesis

3.9 Prove that (a) if /(f) is even, its Fourier transform F(ja>) is also an even
function; (6) if /(/) is odd, its Fourier transform is odd and pure imaginary.

F(ju)

+0

(i)

PROS.
3.10

Find the inverse transform of

F(jm)

shown
domain?

as

3.10

in the figure.

= P 6(fo -

What can you

Wo)

+ P 8{(o +

eo )

say about line spectra in the frequency

chapter

Differential equations

INTRODUCTION

4.1

This chapter

is

devoted to a brief study of ordinary linear differential

We will concentrate on the mathematical aspects of differential

equations.

equations and leave the physical applications for Chapter

equations considered herein have the general form

5.

The differen-

tial

/MO, *'('),
where
t.

*< B >(0,

t]

(4.1)

the independent variable and x(t) is a function dependent upon

superscripted terms >(/) indicate the rth derivative of x(t) with

t is

The

respect to

t,

namely,
>/a
x(t)

d(i)

x(t)

~df~

(42)

F
in Eq. 4.1 is x(t) and must be obtained as an
of /. When we substitute the explicit solution x(t) into F,
the equation must equal zero. If fin Eq. 4.1 is an ordinary
linear differThe

solution of

explicit function

ential equation,

it is

(B>
* (0

given by the general equation

+ a^x^Xt) +

+ , x\t) + oo x(t) =f(t) (4.3)

The order of the equation is n, the order of the highest derivative
term.
The term/(0 on the right-hand side of the equation is Xht forcing function
or driver, and is independent of x(i). When /(f) is identically
zero, the

equation is said to be homogeneous; otherwise, the equation is

nonhomogeneous.
In this chapter we will restrict our study to ordinary, linear
differential
equations with constant coefficients. Let us now examine the
meanings of
these terms.
75

Network

76

Ordinary.

anal/sis

and synthesis

An ordinary differential equation is one in which there is only

one independent variable

our case,

(in

t).

As a result there is no need

for

partial derivatives.

conConstant coefficients. The coefficients a n , o^, ...,a alt a are

stant, independent of the variable t.
the
differential equation is linear if it contains only terms of
Linear.
For
Eq.
4.3.
by
given
as
derivatives,
first degree in x(t) and all its higher

example, the equation

3x'(t)

is

a linear

+ 2z(0 = sin

On the other hand,

differential equation.

3[*

+ 2*(f) x\t) + 4x(0 =

nonlinear, because the terms [x'(t)]*

is

(4.4)

and

5t

x(t) x'(t) are nonlinear

(4.5)

by the

definition just given.

important implication of the linearity property is the superposition

and x^t) are
x
property. According to the superposition property, if x (0
and/,(0,
functions/^)
forcing
for
equation
solutions of a given differential
of
combination
linear
were
any
function
then, if the forcing

An

respectively,

as
JUt) and/,(0 such

AO-afM + btff)

(4-6)

the solution would be

x(r)-ai(0

(4 - 7 )

+ &*40

that the
where a and b are arbitrary constants. It should be emphasized
kept in
be
should
and
important
extremely
is
superposition property

mind
4.2

in

any discussion of linear

differential equations.

HOMOGENEOUS LINEAR

DIFFERENTIAL EQUATIONS

homogeneous,
This section deals with some methods for the solution of
let us find
First,
linear differential equations with constant coefficients.
the solution to the equation
x'(t)

- 2*(<) =

(4.8)

Now, with a little prestidigitation, we assume the solution to be of the form

*(0
where
is

C is any arbitrary constant.

truly a solution of Eq. 4.8.

we

Ce"

Let us check to see whether x(t)

(4.9)

Ce2 *

Substituting the assumed solution in Eq.

obtain

2Ce* f

- 2Ce* =

tA

4.8,

tn

(4.10)

77

Differential equations
It

can be shown, in general, that the solutions of homogeneous, linear

equations consist of exponential terms of the form C<e p <*. To

differential

vt
obtain the solution of any differential equation, we substitute Ce for x(t)
for
which
the
equation
is
in the equation and determine those values ofp

zero. In other words, given the general equation

an *<">(*)

we

let x(t)

+ +

Ce", so that Eq.

4.11

+ a, x(f) =

! x'(t)

(4.1 1)

becomes

(4.12)
+ a# + a,) =
+ a^tf*- +
pt
Since e cannot be zero except at/? = oo, the only nontrivial solutions
1

Ce'Kanp"

when the polynomial

for Eq. 4.12 occur

H(p)

- <W" + a-tf"-1 +

'

W+

(4.13)

Equation 4.13 is often referred to as the characteristic equation, and is

denoted symbolically in this discussion as H(p). The characteristic equation
is zero only at its roots. Therefore, let us factor H(j>) to give
H(j>)

= an(p- PMp- Pi)---(p- p^

(4.14)

solutions
t .... Cw_ 1c*-' are all
of Eq. 4.11. By the superposition principle, the total solution is a linear
combination of all the individual solutions. Therefore, the total solution
of the differential equation is

From Eq. 4.14, we note that C^e"**, Cxe***

x(t)

Q**'

<:_!-

(4.15)

, C
,
x,
B_i are generally complex. The solution x(t) in Eq.
, Cn_x are uniquely
not unique unless the constants C, Cls
specified. In order to determine the constants Cit we need n additional
pieces of information about the equation. These pieces of information are
usually specified in terms of values of x(t) and its derivatives at t
0+,

where

C C

= CV' +

4. IS is

and are therefore referred to as initial conditions. To obtain ft coefficients,

"-1
, a;
>(0+). In a number
we must be given the values <r(0+), x'(0+),
(

of special cases, the values at

/

= 0+.

x'(0 ),

to solve

t = 0

are not equal to the values at

*(0 ),
=
values
at
In
order
determine
the
f
0+
we
must
^"-^(O ),
for the constants C{
This problem arises when the forcing
If the initial specifications are given in terms of

an impulse function or any of its derivatives. We will

discuss this problem in detail in Section 4.4.
For example, in Eq. 4.9 if we are given that *(0+) = 4, then we obtain
the constant from the equation
function /(f)

is

x(0+)

= Ce*=C

(4.16)

Network

78

so that x(t)

is

analysis

and synthesis

uniquely determined to be
x(t)

Example

Find the solution for

4.1

*'(/)

given the

= 4e*

initial

5x\t)

+ 4a</) =

(4.17)

conditions

a<0+)-2

= -1

x'(0+)

From the given equation, we first obtain the characteristic equation

Solution.

= p* +

H(p)
which factors into

+ 4X/> +

(/>

5p

1)

(4.18)

(4.19)

The roots of the characteristic equation (referred to here as

step = -1 p = 4. Then x(t) takes the form

characteristic values)

= Qe-* +

x{t)

From

the initial condition

(0+)

x(0+)

2,

we

(4.20)

C^e-**

obtain the equation

Q+C

=2 =

(4.21)

In order to solve for Cx and C, explicitly, we need the additional initial condition
x'(0+) = -1. Taking the derivative of x(i) in Eq. 4.20, we have

= -Qe-' - 4Q*-4

X '(f)

Atf =0+,a;'(0is
a:'(0+)

Thus the

final solution is

Next,

we examine

multiple roots.

root p

=p

(4.23)

we find

that

=i

a*/)

the case

(4.22)

-Q -4C

= -1 =

Solving Eqs. 4.21 and 4.23 simultaneously,

Ci

C% = J
= %r* - \e~**

when

Specifically, let

(4.24)

the characteristic equation H(p) has

us consider the case where H(p) has a

of multiplicity k as given by

H(p)

= an(p -

k
)

(p

-pd---(p-pn)

(4-25)

be left to the reader to show that the solution must then contain k
terms involving e" * of the form

It will

x(i)

CV"

+ C01 te*>* +

+
+ C^*** +

Cnt'e***

+ C^t*-^"*
Cje "' +
+ Cne*'*

(4.26)

where the double-scripted terms in Eq. 4.26 denote the terms in the solution
due to the multiple root, (p #,)*

79

Differential equations

Example 4.2

Solve the equation

*(/)

- Sx'(t) +

16*(r)

a<0+)2 and

with
Solution.

The characteristic equation

H{p) ~p*
Since H(p) has a double root at/>
x(t)

In order to determine

=4,

final solution is

"=4

is

16

= (p - 4)

(4.28)

the solution must take the form

and C we evaluate

a<r)

x(f)

(4.29)

and

'(') at r

= 0+

to give

= C1 =2
(4.30)

=4Q +C =4

*'(<>+)

Another

*'(<>+)

(4.27)

C^* + Cy*"

(0+)

Thus the

ip

= 2c*' - 4te*'

interesting case arises

(4.31)

when H(p) has complex conjugate

roots.

Consider the equation

H{p)

- a (p -pMp-pJ

(4.32)

where />! and />, are complex conjugate roots, that

Px,Pi

The

solution x(t) then takes the

x(t)

Expanding the term

x(t)

e*"

is,

ojo>

(4.33)

form

Ci<*W">*

<V'>'

by Euler's equation,

= C e*(cos cot +j sin cat) +

t

x(t)

(4.34)

can be expressed as

Cje^cos

cor

y sin cat)

(4.35)

which reduces to
x(t)

= (Q +

C,y* cos

cor

+j(C1

Let us introduce two new constants,

more convenient form

CJe'*

sin cor

and Af so that

(4.36)
x(r)

may be

expressed in the

x(t)

where 3/x and

= Mje'* cos cor + Aftf"' sin cat

are related to the constants

M! =

Cx and C2 by the equations

d + C,

^. =y(C! - C2)
The

constants Afj

initial

conditions.

and

(4.37)

are determined in the usual

(4.38)

manner from

Network

80

and synthesis

analysis

Another convenient form for the solution x(t) can be obtained if we

and <f>, defined by the equations
introduce still another pair of constants,

Ml = MsiD *

= M cos

we

obtain another form of x(i), namely,

With the constants

M and

<f>

+ 2*'(/) + 5x(t) =
=

a<0+)

The

*'(<>+)=

H(p)

characteristic equation

+ 2/7 +

H(p) =/>*

= (p +

is

+j2Kp +

x\t) is

The

= 0+

a>

2.

(4.43)
(4.44)

= Mx( -"' cos 2/ - 2e"' sin 20 + M -e~* sin 2/ + 2c" cos 2t)

(4.45)

+, we obtain the equation

'(<)+)

x(t)

we had used

+ 2MS

-Afj

Solving Eqs. 4.44 and 4.46 simultaneously,

Thus the final solution is

If

(4.42)

derivative of x(i) is

*'(*)

At

-j2)

= M&-* cos 2t + Mtf-* sin It

*(0+) - 1 - Mt

x(t)

At

we have a = -1 and

so that, assuming the form of solution in Eq. 4.37,

Then

(4.41)

conditions

initial

Solution.

(4.40)

<f>)

Solve the equation

x\t)
with the

<f>

- MC* sin (a>t +

x{i)

Example 4.3

(4.39)

<r*(cos 2/

we

find

(4.46)

and

+ sin 2t)

the form of x(t) given in Eq. 4.40,

M, =

+$.

(4.47)

we would have

obtained the

solution
x(t)

Vfe-* sin

[2t

tan"1 (2)]

Now let us consider a differential equation that

have discussed concerning characteristic values.
Example 4.4

The

*<(/)

The

initial

differential

+ 9*

(4

>(f)

equation

(4.48)

illustrates everything

is

+ 32x<(0 + 5Sx'(t) + 56<r'(0 + 24a</) =

conditions are
a!

U)(0+)=0
x'(0+)

= -1

5c

(0+)

x'(0+)

(S,

we

a<0+)

(4.49)

Differential equations

The characteristic equation

Solution.

H(p)

81

is

= p* + 9/ + 32/> + SSp* + 56p + 24 -

(4.50)

which factors into

H(p)

- (p +

+yi)0

From H(p) we immediately write x(t)

a<0

- Mi*-

cos

-;1)0>

+ 2)ty +

(4.51)

as

+ M^r* sin f + Qc-** +

Since there are five coefficients,

3)

Qte"**

+ Qe-3

(4.52)

we need a corresponding number of equations

to evaluate the unknowns. These are

a<0+)
a;'(0+)

x'(0+)
a;)(0+)
*<((> +)

- Mx + C + C8 = 1
- -Mx + Mt - 2C - 3Ct + Q - -2M, + 4C - 4Q + 9Ct = -1
= 2Mj, + 2Mt - 8C + 12Q - 27C, =
= -4M1 + 16C + 81C, - 32CX =

Solving these five equations simultaneously,

Mi -

so that the final solution

C =

-1

(4.53)
1

we obtain

Q -i

=0

C.

is

a</)

- $e~' sin /+-*+ Jte-**

(4.54)

We have seen that the solution of a homogeneous, differential equation

may

take different forms depending

upon

the roots of

its

characteristic

equation. Table 4.1 should be useful in determining the particular form of

solution.

TABLE

4.1

Forms of Solution

Roots of H(p)
1.

2.
3.

Single real root, p

=p

Root of multiplicity, k, (p -*)*

Complex roots at/>, , = a jm

e p,t
Ce*i*

Cje'i*

Me* sin (mt

4.

Complex roots of multiplicity k

*tP*.6 = " j<>>

M C* cos mt + Mge"*

+ CfcV'-W

sin ait

+ +)

M^* cos mt + Mjte"* cos / +

+ Aft-if*-1**' cos mt + NtfP* sin mt
+ Nite?* sin mt H
+ Nk_1tk- sin mt

Network

82

4.3

analysis and synthesis

NONHOMOGENEOUS EQUATIONS

a nonhomogeneous
equation is one in which the forcing function f(t) is not
identically zero for all t. In this section, we will discuss methods for
obtaining the solution x(t) of an equation with constant coefficients

As we mentioned in the introduction to this chapter,

differential

an *<>(*)

+ a^ x-(0 +

+a

=f(t)

x(t)

(4.55)

Let x p (f) be a particular solution for Eq. 4.55, and let x e (t) be the solution
in Eq. 4.55.
of the homogeneous equation obtained by letting f(t)

It is readily

seen that

<t)
is

(4-56)

also a solution of Eq. 4.55. According to the uniqueness theorem, the

solution x(t) in Eq. 4.56

is

= 0+. 1

mentary function; and x(t)

nonhomogeneous

the unique solution for the

equation if it
In Eq. 4.56, x p(t)

differential
t

= xjf) + x (0

satisfies
is

is

the specified

initial

the particular integral; x a(t)

conditions at
is

the comple-

the total solution.

Since we already know how to find the complementary function x e (t),

we now have to find the particular integral x p(t). In solving for xp(t), a
very reliable rule of thumb is that x p(t) usually takes the same form as the
forcing function if/(0 can be expressed as a sum of exponential functions.
Specifically,

example,

xp{t) assumes the form of /(/) plus all its

= a sin cot, then xjf) takes the form

For

derivatives.

if/(f)

x P(t)

= A sin cot + B cos cot

that must be determined are the coefficients A and B

of the terms in xjf). The method for obtaining x p(t) is appropriately
called the method of undetermined coefficients or unknown coefficients.

The only unknowns

In illustrating the method of unknown coefficients,

let

us take/(0 to be

/(0=<*e"
where a and

/S

is

the

then assume x p (t) to have a

is,

xjf)

and

We

are arbitrary constants.

similar form, that

(4.57)

unknown

coefficient.

= Ae"
To

(4.58)

determine A,

we simply

substitute

the assumed solution x p(t) into the differential equation. Thus,

AefiXaJ"

a,,.^"-

+ aj +

= oe"

(4.59)

See, for example, C. R. Wylie, Advanced Engineering Mathematics (2nd

McGraw-Hill Book Company, New York, 1960, pp. 83-84.

ed.),

Differential equations

83

We sec that the polynomial within the parentheses is the characteristic

equation H(p) with /
Consequently, the unknown coefficient is
/?.
obtained as

A = -^provided that H(fi) jt

Example 4.5

0.

Determine the solution of the equation

+ 3x'(t) + Ttff) - 4e*

**(r)

with the

initial conditions,

Solution.

The

a<0+)

1,

x'(0+)

= 1.

+ 3/> + 2 = (p + 2Xp +

so that the complementary function

*JLi)

For the forcing function /(f)

1)

is

= cie-* + c*-*

4e, the constants in Eq. 4.60 are

= 4,

=> 1.

^ = _i_ = ?

Then

H(l)

Thus we obtain

xp(t)

= |e*

total solution is

*(')

To

(4.61)

characteristic equation is

Hip) -/>

The

(4.60)

- *c(0 + *(0 = Cxe- + Cir" + |e

evaluate the constants

Q and C we substitute the given

(4.62)

initial conditions,

namely,

40+) =
*'(0+)
Solving Eq. 4.63,

we find

"Q+Cg+f
(4 ' 63)

= -1 - -Ci - 2C, + f

that

Q=

a</)

. _ e-' +

1,

C,

= $.

Consequently,

+ fe*

(4.64)

we solve for the constants x and t from the

conditions for the total solution. This is because initial conditions are
not given for xe(t) or *(/), but for the total solution.
It

should be pointed out that

initial

Next,

let

us consider an example of a constant forcing function/(f)

4.60 if we resort to the artifice

= a.

We may use Eq.

/(r)

=a=

ae*

(4.65)

Network

84
that

we

is, j8

= 0.

analysis

For the

and synthesis

differential

equation in Example 4.5 with/(/)

*(0

and

= Cltr* + C*r + 2

(4.67)

the forcing function is a sine or cosine function,

consider the forcing function to be of exponential form and

When

the

= 4,

see that

method of undetermined
fit)

coefficients

and Eq.

4.60.

we can still
make use of

Suppose

= oe* = a(cos <ot + j sin oat)

(4.68)

then the particular integral x vlit) can be written as

x Plit)

From the

Re

MO + J Im MO
we can show

superposition principle,

=
fit) =
fit)

if

if

o>t

then

a sin tot

then

a.

cos

(4 ' 69)

that

= Re MO
*,(0 = Im * rt(f)
x(0

mt or a
Consequently, whether the excitation is a cosine function a cos
iat
<xe
driver
exponential
;
an
fit)
use
can
we
or,
sine function a sin
integral.
particular
resulting
the
of
part
imaginary
or
real
the
then we take

Find the particular integral for the equation

Example 4.6

x'it)

Solution.

First, let

+ 5*'(0 + 4a<<) = 2 sin 3*

(4.70)

us take the excitation to be

/i(0

= 2e*

(4.71)

so that the particular integral *.(/) takes the form

rt(0

- Ae*

(4.72)

From the characteristic equation

H(p)

we determine

the coefficient

H(j2)

Then

MO

=/>*

to be

_
-5 +jlS

=,

L^tan-Hs)--]
5 VlO

1^ o^ ^*2

is

+5p +4

* -'5VI0

(4.73)

474>

<

Differential equations

and the

particular integral x(t) for the original driver /(f)

85

= 2 sin 3/ is

- Im xpl(t) = -= sin [3f + tan"1 (3) 2

xjf)

5V10

*]

(4.75)

There are certain limitations to the applicability of the method of

undetermined coefficients. If /(f) were, for example, a Bessel function
7 (f)> we could not assume xv(t) to be a Bessel function of the same form
(if it is a Bessel function at all). However, we may apply the method to
forcing functions of the following types

= A; constant.
= A(t n + Vi' +
+ V + b );
pt
3. /(f) = e
real
or
complex.
p
!

/(0

2. /(f)

'

n, integer.

4.

Any function formed by multiplying terms of type

For the purposes of

linear

network

analysis, the

1, 2,

method

or

3.

more than

is

adequate.

Suppose the forcing function were

/(f)

The

particular integral

xp(t)
where the

4.4

AND

In this section

p = a+jm

can be written as

= (AJ* +

coefficients

STEP

= At*e pt

A& A_lf

4^ +
.

. ,

...

A A
lt

+ A 1t + A )e*

(4.76)

are to be determined.

IMPULSE RESPONSE
we

will discuss solutions of differential equations with

step or impulse forcing functions. In physical applications these solutions

are called, respectively, step responses and impulse responses.

quantities, the step

and impulse responses of a

As

physical

linear system are highly

of system performance. In Chapter 7 it will be shown

is given by
its impulse response.
Moreover, a reliable measure of the transient
behavior of the system is given by its step and impulse response. In this
section, we will be concerned with the mathematical problem of solving
for the impulse and step response, given a linear differential equation with
significant measures

that a precise mathematical description of a linear system

initial

conditions at

From Chapter 2,

= 0.

recall that the definition

(f)

=
=

^
f <
f

of the unit step function was

Network

86

analysts

and synthesis

and the unit impulse was shown to have the properties:

d(t)

oo

=
and

tj*0
<3(0 df

JoJo-

in addition,

we have

the relationship
d(t)

= i*@
dt

As

the definitions of d(t) and u(t) indicate, both functions have dis-

continuities ait

0.

In dealing with initial conditions for step and impulse

we must then recognize that the

x"(f), etc., may not be continuous at t

drivers,

solution x{t)

x'(t),

= 0.

and

its

derivatives

In other words,

may be

it

that
*<>(()-) ji z<>(0+)

x (-i)(0_)
x(0-)
In

many

^ a;<"-"(0+)
5*

*(0+)

physical problems, the initial conditions are given at

However, to evaluate the unknown

have the initial conditions at t = 0+. Our

constants of the total solution,

conditions at

is

= 0

we must

task, then, is to determine the

= 0+, given the initial conditions at = 0

discussed here

The method

borrowed from electromagnetic theory and

is

often

referred to as "integrating through a Green's function."*

Consider the

differential

a n *<>(t)

equation with an impulse forcing function

+ a*.,. xt-\t) +-- + ax(f) = A d\t)

(4.77)

To

insure that the right-hand side of Eq. 4.77 will equal the left-hand
x0) must contain an impulse.
one of the terms z (B, (f), * (n_1) (0>
The question is, "Which term contains the impulse ?" A close examination
shows that the highest derivative term x ln) (t) must contain the impulse,
because if * <n-1) (f) contained the impulse, x (B) (f) would contain a doublet
C d'(t). This argument holds, similarly, for all lower derivative terms of
*-u
(n>
(f) contains the impulse, then s
x(i). If the term 3
(0 would contain a
{n~*
that,
We
conclude
therefore
step and x
for an impulse
(0> a ramp.
at i = 0.
discontinuous
derivative
terms
are
highest
the
two
forcing function,
side,

'The Green's function is another name for impulse response; see, for example,
Morse and Feshbach, Methods of Theoretical Physics, McGraw-Hill Book Company,

New York,

1952, Chapter 7.

87

Differential equations

For a step forcing function, only the highest derivative term is discontinuous
at

0.

0 , our task is to
t
determine the values a; (B) (0+) and * <n - 1, (0+) for an impulse forcing
function. Referring to Eq. 4.77, let us integrate the equation between
Since initial conditions are usually given at

= 0

and

)xM

an

(t)

dt

Jo-

AP

==

0+, namely,

+ a^ f ^'-"(O dt +
Jo-

a ]x{t)dt
Jo-

= A f %*)<*'
Jo

(4-78)
.

After integrating,

aB [*(-(0+)

we

obtain

- *<-i>(0-)] + a-i[* (B-

a,

- *<-(0-)] +

(0+)

=A

(4.79)

We know that all derivative terms below (n

are continuous at

1)

= 0.

Consequently, Eq. 4.79 simplifies to

an [x<-1 >(0+)

xl

so that

- ^"(O-)] = A

^v(0+) = + * *-"(0-)
(

We must next determine * (n> (0+).

in Eq. 4.77

At

all

(4.81)

= 0+, the differential equation

is

+ a^ *<"-(0+) + + Oo *(0+) =
derivative terms below ( 1) are continuous, and

an *<">(0+)
Since

(4.80)

(4.82)

since

we

have already solved for z(B-1) (0+), we find that

*<">(<)+)

- - -- [0^x^X0+) +

ai x'(0+)

*(0+)]

flo

<*n

(4.83)

For a step forcing function A u(t), all derivative terms except x<">(f), are
(B)
continuous at t
0. To determine a;
(0+), we derive in a manner
similar to Eq. 4.83, the expression

*<>(0+)

= - - -

The process of determining

is

-1>

[a^1 x(B
initial

(0+)

conditions

ao*(0+)]

when the

an impulse or one of its higher derivatives can be

(4.84)

forcing function

simplified

by the visual

Network

88

shown in Eqs.

process
its

analysis

and synthesis

and 4.86. Above each derivative term we draw

Note that we need only go as low

4.85

associated highest-order singularity.

as a step in this visual aid.

OnX^Kt)

ax<\t)
It

oB_1a: (B 1 W

OaV- ^) + (Wf^O +
1

should be noted that

if a derivative

doubletit

for example, a

+ a^^-'KO + + ao^O =

<*'(')

+ flo^W = *0

(4.85)

(4.86)

term contains a certain singularity

also contains all lower derivative terms.

For

example, in the equation

*'(/)

we assume

3'(0

2a</)

(4.87)

4<5'(/)

the following forms for the derivative terms at

= 0:

= A S'(t) + B 8{f) + C (0
x'(f) - A 8(t) + B u(t)
x(t) = A u(t)

*"(0

Substituting Eq. 4.88 into Eq. 4.87,

A d'(t) + B (5(0 +
Or

in a

Cu(t)

we

(4.88)

obtain

+ 2A u(f) - 4d'(t)

(4.89)

+ (C + 3B + 2A) u(0 - 4d'(0

(4.90)

+ I A 8(t) +

3B (r)

more convenient form, we have

A d'(t) +
Equating

(B

3A)

like coefficients

<J(0

on both

sides of the Eq. 3.90 gives

A=

=
C + 3B + 2A =
B+

3A

4
(4.91)

89

Differential equations

from which we obtain

B = 12

C=

and

Therefore, at

28.

= 0,

it is

true that

= M(t) - 12(5(0 + 28(0

x'(0 = 4d(0 - 12(0
*(*) = 4u(t)

**(*)

The w(0 terms

conditions at

we

in Eq. 4.92 gives rise to the discontinuities in the initial

= 0. We are given the initial conditions at t = 0

evaluate the A, B,

conditions at

example,

(4.92)

Once

C coefficients in Eq. 4.88, we can obtain the initial

= 0+ by referring to the coefficients of the step terms.

For

if

= -2
*'(0-) - -1
*(<)-) = 7
*(<)-)

Then from Eq. 4.92 we obtain

x(0+)

=-2 + 4 = 2

= -1 - 12 ex -13
**(()+) = 7 + 28 = 35
x\0+)

(4.93)

The total solution of Eq. 4.87 is obtained as though it were a homogeneous

equation, since d'(t) =
for t j*s 0. The only influence the doublet driver
has

is

to produce discontinuities in the initial conditions at i

evaluated the

initial

conditions at

= 0+,

we can

= 0.

Having

obtain the total

solution with ease. Thus,

z(0

From

Eq. 4.93

we

= C <r* +
x

C*tr

(4.94)

readily obtain

*(0

= (-9er* +

ll<r-) u(t)

(4.95)

The total solution of a differential equation with a step or impulse forcing

For a step
is obtained in an equally straightforward manner.

function

forcing function, only the highest derivative term has a discontinuity at

t = 0. Since we do not need the initial condition of the highest derivative
term for our solution, we proceed as if we were solving a standard nonhomogeneous equation with a constant forcing function. For an impulse
driver, once we determine the initial conditions at t = 0+, the equation is
solved in the same manner as a homogeneous equation.

90

Network

Example 4.7

analysis

and synthesis

Find the step and impulse response for the equation

+ 4x'(t) +

2x"(f)

where /(f)

<5(f)

and/(f)

I0x(0 =/(f)

(f), respectively.

The

initial

conditions at

are

a<0-)
Solution.

Let us

contains an impulse

a ramp, and

we use Eq.
*'(0

Note that we

x"(0-)

find the impulse response.

and a step

note that the x" term

Thus x(0+)

0.

x term contains

= x(0 ) = 0. To

4.81

- - + *'(0 -) -

+)

=\

a2

actually need only x(0

+) and

+) to evaluate the constants

we proceed to the comple-

x'(0

Next,

mentary function x e(t). The characteristic equation

- 2(p* + lp +

- lip +

5)

is

+jlKp +

(4.96)

2.

2.

for the second-order differential equation.

H(p)

We

the x' term contains a step ; the

therefore continuous at

is

obtain a;'(0+),

first

*'(0-)

Since H(p) has a pair of complex conjugate roots,

-j2)

we use a standard form

= Me~* sin (2f + #

Substituting the initial conditions at f = 0+, we obtain
x c (i)

a<0+)
x'(0+)

from which we find ^

denote here as xt(t), is

(4.98)

= J = 2Mcos $ -

M=

**) =

\.

(4.99)
sin

<f>

Thus the impulse response, which we

1~* sin 2f u{i)

Next we must solve for the step response xu{t). For convenience,
complementary function as

The

particular integral

constant /(f)

is

(4.100)
let

= e-KA^ sin 2f + A t cos 2f)

us write the

(4.101)

evaluated by considering the forcing function a

so that
a,

The

for

=0 = Afsin^

and

xe(t)

(4.97)

jt) _!_ = J-

(4.102)

total solution is then

a<f)

- (A t sin 2f + At cos 20*?-* + ^

(4.103)

Since x'(t) and x(t) must be continuous for a step forcing function,
a

<0+)=a<0-)=

a;'(0+)-a; , (0-)

=0

(4.104)

Differential equations

Substituting these initial conditions into x(i)

and *'(/), we find that Ax
-0.05 A t
-0. 1 .

Therefore, the step response

*(*)=

d/dt

C*

cos 2,)]*)

WO

L-

*,>

(4.105)

>*,(/)

FIG.

Note

4.1.

that the impulse response

step response are related

and the
by the equation
x,

(0

i.
*(<)
at

(4.106)

Wecan demonstrate Eq.4.106by thefollowingprocedure.

xu(t)

> *j(0
C*

L_

--*(0.5sin2/+

0.1[1

d/dt

> *(0

*,(')

is

91

Let us substitute

into the original equation

Differentiating both sides,

_|

.(*)]

from which Eq. 4.106

Generalizing,

equation,

we

+ 4 ^ x.(0 +

.(*)

. h(0

10*(0

(4.107)

we have

+ 4 1[| *(o] +

io[^ *(*)]

<H0

(4.108)

follows.

see that, if we have the step response for a differential

we can

obtain the impulse response by differentiating the step

can also obtain the response to a ramp function f(t)
A p(t) (where A is the height of the step) by integrating the step response.
The relationships discussed here are summarized in Fig. 4.1.
response.

We

INTEGRODIFFERENTIAL EQUATIONS

4.5

In this section,

we

an

will consider

integrodifferential equation of the

form
a *"(t)

a,.!

*-\t)

x(t)

a_t f x(r) dr

= /(/)

(4.109)

Jo

where the coefficients {a, a1K.1

, a_ t } are constants.
In solving an
equation of the form of Eq. 4.109 we use two very similar methods. The
first method is to differentiate both sides of Eq. 4.109 to give
,

a n x (n+1 \t)

+ a^ xM (t) +

The second method

consists

a x\t)

a_t

of a change of variables.

x(t)

-/'()

(4.U0)

We let y'(f) = x(t);

Eq. 4.109 then becomes

a, y(n+1)(t)

(n

*_! y \t)

y'(t)

a_x

y(t)

= f(t)

(4.111)

Network

92

analysis and synthesis

Note that from Eq. 4.110 we obtain x(t) directly. From Eq. 4.111, we
obtain y(t), which we must then differentiate to obtain x(t). An important
point to keep in mind is that we might have to derive some additional
the
initial conditions in order to have a sufficient number to evaluate

unknown

constants.
Solve the integrodifferential equation

Example 4.8

x'(t)

+2f

3a<0

*(t)

dr

(4.112)

5(0

is x(0 ) = 1.
Since the characteristic equation of Eq. 4.112 is of second degree,
obtain *'(<)+) from the
need an additional initial condition x'(0+).

The

initial

condition

Solution.

we

We

given equation at

= +

+ 3<0+) + 2
Since x(t) is continuous at / = 0,

a< T) dr

*'(0+)

(4.113)

<H

a<T)</T=0

(4.114)

(0+)-a<0-)-

(4.115)

I
and

x'(0 +)

Therefore,

-Method

1.

The complementary function

+ 3*'(0 + 2a<0 =
is

initial

conditions for

= Car* +

a<0+) and

x(t)
2.

(4.1 16)

Letting y'(t)

becomes

We know that

obtain
(4.117)

5<K0

the

MO +

y'(P+)

C*r*>

*'(<>+),

- 4e~' -

= x(t),

y"(0

we

then

*.(0

Method

- 3s(0 +) = 2

Differentiating both sides of Eq. 4.112,

x"(t)

Using the

(4.118)

we

obtain the total solution

(4.119)

3e-*

original

2y(t)

= (0+) -

differential

equation then

5(0
1

(4120)
,,,,

(4.121;

j,'(0+)= *'(<>+)= 2

From Eq.

4.120, at t

= 0+, we obtain

y(0 +)

Without going into

i[5

- y'(0 +) - 3y'(0 +)] -

details, the total solution

j<0

(4.122)

can be determined as

- -4e- + |e-* + \

(4- 123 >

93

Differential equations
Differentiating y{t),

we

obtain

<i)

4.6

= y'(f) = 4e~' - 3<?-*'

(4.124)

SIMULTANEOUS DIFFERENTIAL EQUATIONS

Up to this point, we have considered only differential equations with a

single

with

dependent variable

In this section,

x{t).

more than one dependent

variable.

we

will discuss equations

We shall limit our discussion to

equations with two unknowns, x(t) and y(t).

The methods

described here,

however, are applicable to any number of unknowns. Consider

system of homogeneous equations

+ a, x(0 + & y'(t) + y(t) y'(0 + \ y(t) = o

yi *'(0 + Yo *(0 +

x *'(0

first

the

fi,

(4.125)

<$!

where a

f} it y it d i are arbitrary constants. The complementary function

obtained by assuming that

so that the characteristic equation

H(p)

is

given by the determinant

+
+ Yo>

(a^

oto)

(ftP

is

+ A)
(<V + to)

<fc/>

(4.126)

The

roots of H(j>) are found by setting the determinant equal to zero,

that

is,

- OV + PoXYiP + Vo) =
It is seen that a nontrivial solution of H(p) =
exists only if
(tf>

(*

)(V

*)

+ KV +

<>a)

* OV + &XftP + y<0

(4.127)

(4.128)

Assuming that the preceding condition holds, we see that H(p) is a seconddegree polynomial tap and can be expressed in factored form as
H(p)
where

= C(p- Po)(p - />0

C is a constant multiplier.

The complementary

= K^e** + K***
y(t) + Kf
x(t)

(4.129)

functions are

(4.130)

Network

94

anal/sis and synthesis

and the constants

K K K Kt are determined from initial conditions.

t,

x,

9,

As in the case of a single unknown,

tf Po

If

/>i

if H(p)

has a pair of dduble roots;

i.e.,

then
x(t)

(#,

y(t)

(tf,

+ Kt t)e*<*

K^e***
(4.131)

H(p) has a pair of conjugate roots,

Pi

= a jto

Pi*)
then

x(t)

M^e"*

sin (cot

<^)

y(f)

Mte" sin (tor

<f>J

(4.132)

Consider the system of equations

Example 4.9.

2x'(t)

+ 4x(t) + y'(t) -y(f)=0

(4.133)

x\t)

with the

initial

2x(0

+ y'(0 + y(t) =

conditions

*'(0+)2

y '(o+)=-3

a<0+)=0
Solution.

The

(4.134)
1

characteristic equation is

Evaluating the determinant,

#0)
so that

i/(0+)

+4

+2

we find

that

=/>*

5/>

(4.135)

+ 6 = (p + 2Xp + 3)

y{t)

a</)

- KtfT* + J>-

Kjfi-"

+ A^-8

(4.136)

'

(4.137)

With the initial conditions, z'(0+) - 2,a<0+) = 0, we obtain K, =2,Kt = -2.

From the conditions y'(0+) - -3, y(0+) = 1, we obtain Kt = 0, JCg = 1.
Thus the final solutions are
(4.138)
y(.t)

-s

Differential equations

Next,

95

us determine the solutions for a set of nonhomogeneous

We use the method of undetermined coefficients
here. Consider first an exponential forcing function given by
the set of
equations
let

differential equations.

+ aye + piy + p = Ne
Yi*' + y<P + b y' + d& =
We first assume that
xp(t) = Ae"
alX '

'

iff

(4.139)

yv(t)

(4.140)

= Be9

Then Eq. 4.139 becomes

+ xJA + (^0 +
(.YJ + YM + O*i0 +

Qtfl

The determinant

for the set of equations 4.141

H(8)

+
+ Y*

i0

= A(0) =

where H(6)

is

(4.141)

is

PiO

<,

YiO

=N
d )B =
o)B

the characteristic equation with p

+ Po
+\
0.

(4.142)

We now determine

#4he undetermined coefficients A and B from A(0) and its cofactors, namely,

A=

*A

1X (0)

A(0)
(4.143)

JVA ia(e)
B=
A(0)

where

A is

Example

the i/th cofactor of A(0).

4.10.

Solve the set of equations

2x'

+ Ax + y' - y = 3e**
(4.144)

x'+2x+y'+y=0

= 1, a<0+) = 0, y'(0+) = 0, y(0+) = -l.

The complementary functions *c(f) and ye(i), as well as the characequation H(p), were determined in Example 4.9. Now we must find

given the conditions x'(0+)

Solution.
teristic

A and B in the equations

The

*,(/)

- Ae**

y,(t)

- Be**

characteristic equation withy*

(4.145)

= 4 is

2(4)

H(A)~
(4)

+4
+2

(4)-l
(4)

42

(4.146)

Network

96

analysis

and synthesis

Then we obtain from Eq. 4.143 the constants

A The incomplete

B - -t

solutions are
*(f)
y(t)

= KiT* + Ktfr** + &e

- K&-* + Kie-* - fe"

(4.147)

Substituting for the initial conditions,

we

obtain

finally

s
y(0=K-4" '-34')
(4.148)

Example 4.11.

Solve the system of equations

+ 4x + y' + ly =
x' +x +y' +3y =

lx'

given the

initial

Solution.

First

a<0-)

= *'(<>-) = jKO-) = y'(0-) -

we find

the characteristic equation

A(/,)

2/>

/>

+4 / +

(4.150)

p +

simplifies to give

H(p)

=(pt +2p+5)=(p +

The complementary

and
-'

functions x c (t)

xe(t)
y e (t)

The

(4.149)
56(f)

conditions

J5T(Cp>

which

5(/)

= ^e

-'

= jBiC

j/

(f)

+j2)(p

-j2)

(4.151)

are then

cos It

+ A&-* sin 2/

cos 2/

+ B^r* sin 2f

(4.152)

particular solutions are obtained for the set of equations with t

>

0,

namely,

+ 4x + y' + ly =
*' +
+ y' + 3y -

2x'

5
(4.153)

a:

coefficients, we assume that x v and y,, are

Since the forcing function 5 can be regarded
as an exponential term with zero exponent, that is, 5
5e', we can solve for
a with the use of the characteristic equation H(p) with p = 0. Thus,
x and

Using the method of undetermined

constants:

as,

= Q;

yv

= Ca

#(0)

A(0)

(4.154)

Differential equations

5AU(0)

n
c

Dd

= 5(3)

^--5(or

97

3
(4.155)

c,
The

general solution

is

- ^e-' cos 2r + A^r* sin 2/ + 3

= fl^* cos 2/ + B&-* sin 2r - 1

y(t)

In order to find

The

then

x(t)

y'(0+).

A lt At Blt B* we
,

values for

need the values a<0+), x'(0+), y(0+),

first obtained by integrating the

a<0+) and y(0+) are

0 and f

original differential equations between t

/0+

+ 4* + y' + ly)dt -

0+. Thus,

5(f)<//

Jo-

Jo/0+

(*'

We know

/VH-

(2*'

impulses at

(4.156)

+ x + y' +

3y*<ft

(4.157)

fH-

5d(t)dt

that only the highest derivative terms in both equations contain

f

- 0. Moreover, both x(t) and y(t)

0. Therefore, in the integration

contain, at most, step dis-

continuities at t

r
r

(4a;

After integrating,

+ ly) dt =
(4.158)

(x

3y)rff0

we obtain

2a<0+)+y(0+)=0
(4.159)

*(0+)+y(0+)5
Solving,

we find
a<0+)

To

find x'(0+)

and y'(0+), we

the original equations at

= -5

y(0+)

10

(4.160)

substitute the values for

= 0+.

a<0+) and y(0+) into

Thus,

- 20 + j/'(0+) + 70 = 5
x\0+) - 5 + y'(0+) + 30 =
x'(0+) = -20
y'(0+) = -5

2z'(0+)

so that

Substituting these values into Eq. 4.156,

(4.161)

(4.162)

we eventually obtain the final solutions

x(t)

- ( -%e~* cos It -

y(/)

- (1 \e~* cos 2/ + 3e~* sin It -

14e~* sin It

3) u(t)

(4.163)
1) (f)

Network

98

analysis

and synthesis

Problems
Show

4.1

that

and

3^(0

= Mxe~* cos It

xjj)

m,

M&-* sin It

are solutions for the equation

+ 2x'(t) + 540 =

x'(t)

Show

that

x1

xt

is

also a solution.

Determine only the form of the solution for the equation

4.2

+ 4x'(0 + 340
+ 8x'(t) + 540
*"(/) - 540
x'(0 + 540
x"(t) + 6x'(0 + 2540
z'(0 + 6x'(.0 + 940
conditions 40+) = 1,
**(0

(a)

x"(t)

(6)
(c)

(d)
(e)

(/)
4.3 Given the initial
solutions for

=
=
=
=
=
=
*'(0+)

1, determine the

+ 2540 =
&r'(0 + 16x(t) =
x\t)
x'(t) + 4.8lx'(t) + 5.7640 =

(a)

+
+

x'(t)

(b)
(c)

4.4

Find only the particular

fo'(0

integrals for the equations

x"(t)

(b)
(c)

(d)
(e)

(/)

4.5 Given the initial conditions

solutions for
(a)

x"(t)

(.b)

x"{t)

(c)

*"(/)

4.6

(a)

initial

(b)

'

determine the

+ 4*'(0 + 340 = 5e~* sin It

+ 6x'(t) + 254/) - 2 cos /
+ 8a;'(0 + 1640 = 2

A system is described by the differential equation

y'(0

The

= e-8
= 2 sin 3f
= e~* sin It
= <r"/2
=6
= 2e-* + 3e~at
40+) = 1, x'(0+) =0,

+ 7x'(t) + 1240
x'(t) + 3x'(t) + 240
*'(0 + 2x'(t) + 540
x"(t) + 2*'(0 + Sx(t)
x'(t) + 5.0x'(0 + 6.25x(t)
**(/) + fo'(0 + 540

(a)

3y'(/)

conditions are y(0-)

Given the

differential

1;

- a(0
y'(0-) = 2. Find

2y(t)

equation

as<*>(0

14*'(0

+ 840 -

6<K0

y(0+) and y'(0+).

Differential equations

with the intial conditions *(0-)

12, *'(<>-)
*<*>)+), *'(0+), *'(0+), and a<0+).
4.7 Given the initial conditions x'(0)
for the equations

= 6,

and x"(0-)

= **(0-) = 0,

+ 2x'(/) + 2a</) = 3<K0

+ 7s'(0 + 12a<r) - 5(0
condition x(0 ) = 2, solve

(a)

99

-7, find

find the solutions

x'{t)

x"{t)

(*)

4.8

Given the

initial

the integrodifferential

equations

+ 4 f a<r) </t =

()

x'(t)

(b)

x'(t)

+ 2tft) + 2 f x{r) dr = | e-

(c)

*'(')

+ &<0 + 9 f *(t) oV = 2u(t)

4.9

Given the

with s(0-)
y'(0+).
4.10

set

5a</)

*V) + *(/) + y{t) =

(/)

<K0

*C)+y'(0 + 2y(0
- *'(0-) = y(o-) = y'(0-) = 0,

find *(0+), z'(0+), y(0+),

Solve the set of equations

+ 3*(0 + y'(t) +
*' + *(') + y'W +
All initial conditions at f = 0 are zero.

6y(t)

6y(0

= */)
= (')

Solve the set of equations

+ 2a<0 + y'(0 - y(i) =

+ a</) + y'(0 + 2y(r) =

2s'(0

x'(0

The

of equations

2*'(0

4.11

2 sin

initial

2<5(0

3e~* uSj)

conditions are

a#)-)

*'(0-)=-l

y(0-)=0
j,'(0-)=0

and

chapter

Network

analysis:

5
I

INTRODUCTION

5.1

In this chapter we will apply our knowledge of differential equations to

will assume
the analysis of linear, passive, time-invariant networks.
that the reader is already familiar with Kirchhoff's current and voltage

We

and with methods for writing mesh and node equations for a-c or d-c
1
We will, therefore, consider only briefly the problem of writing
circuits.
mesh and node equations when the independent variable is time t. The
problems in this chapter have the following format: Given an excitation
signal from an energy source and the network, a specified response that
is a current or voltage in the network is to be determined. When relating

laws,

these problems to the mathematics in Chapter 4, we shall see that,

physically, the forcing function corresponds to the excitation ; the network
is

described by the differential equation; and the

is

the response.

unknown

variable x(t)

will be twofold. First, we must write the

problems of the network using Kirchhoff's current and voltage
laws. Next, we must solve these equations for a specified current or
voltage in the network. Both problems are equally important. It is
useless, for example, to solve a differential equation which is set up in-

The problems encountered

differential

conditions are incorrectly specified.

The usual type of problem presented in this chapter might generally be
0, which connects an energy
switch is closed at t
described as follows.

correctly, or

whose

initial

(voltage or current) source to a network (Fig. 5.1).

closing at

is

the energy source whose output

The analog of a switch

is e{t) u(t).

Before the

1
For a comprehensive treatment, see H. H. Skilling, Electrical Engineering Circuits
(2nd Edition), John Wiley and Sons, New York, 1965.

100

Network

analysis:

101

*0

/\

Energy
source'

FIG.

switch

Network

5.1.

Switching action.

closed, the currents

and voltages in the network have known

are the initial conditions. We must then
determine the values of the currents and voltages just after the switch
closes (at / = 0+) to solve the network equations. If the excitation is not
an impulse function or any of its derivatives, the current and voltage
variables are continuous at t = 0. For an impulse driver the values at
/ = 0+ can be determined from methods given in Chapter 4.
Having
obtained the initial conditions, we then go on to solve the network
is

values. These values at

differential equations.

only for
Since

= 0-

Unless otherwise stated,

all

the solutions are valid

^ 0+.

we are dealing only with linear circuits,

it is essential that we bear

the all-important principle of superposition. According to the
superposition principle, the current through any element in a linear circuit

in

mind

with n voltage and

current sources is equal to the algebraic sum of
currents through the same element resulting from the sources taken one
at a time, the other sources having been suppressed. Consider the linear

network depicted in Fig. 5.2a with n voltage and

FIG. 5 Jo

current sources.

102

Network

anal/sis and synthesis

FIG. 5.2b

Suppose we are interested in the current iT(f) through a given element Z,

as shown. Let us open-circuit all the current sources and short-circuit
n 1 voltage sources, leaving only /0 shown in Fig. 5.26. By iVj(t),
we denote the current through Z due to the voltage source /f) alone. In
similar fashion, by iCk(t) we denote the current through Z due to the
current source ik(t) alone, as depicted in Fig. 5.2c.

FIG. 5.2c. Superposition in linear

By

circuits.

the superposition

Network
principle, the total current

algebraic

T(t) due to

of the sources

all

is

103

equal to the

sum

h = \ + '.,+
5.2

analysis:

I.,

,,+

(5.1)

NETWORK ELEMENTS

In this section,

we

will discuss the voltage-current (v-i) relationships

that exist for the basic network elements.

relationships,

it is

important to assign

Before

we examine

these

arbitrary reference polarities for

first

the voltage across an element, and a reference direction of flow for the
current through the element.

assume that the

For the purposes of our

positive polarity for voltage

arrow, as shown by the resistor in Fig.

5.3.

is

discussion,

at the tail

we

of the current

Now, let us review the voltage-

current relationships for the resistor, the inductor, and the capacitor,

which we

first

discussed in Chapter

1.

Resistor

The

resistor

shown

in Fig. 5.3 defines a linear proportionality relation-

ship between v(t) and

i(t),

namely,

K0 - r
i(0

where

m
(5.2)

= Gt*0

R is given in ohms and G in mhos.

Capacitor
For the capacitor shown

in Fig. 5.4a the v-i relationships are

dt
(5.3)

<t)

= -\

i(r)

CJo-

dr

+ o(0-)

-o
'

+
Ut)\

tR

ZZC

**>

v(t)

i(*>\

v(t)

={=C

(a)

FIG.

5.3. Resistor.

FIG. 5.4. (a) Capacitor.

with initial voltage.

(b)
(A)

Capacitor

Network

104

anal/sis and synthesis

J(t)_

<

ml\

v(t)

(J).-)

5.5. (a) Inductor.

C is given in farads.

v(t)

(b)

(a)

FIG.

(6)

Inductor with

initial current.

initial value t> c (0 ) is the voltage across

action. It can be regarded as an
switching
the capacitor just before the
Fig. 5.46. We should point out
shown
in
source,
as
voltage
independent
except impulses and
excitations
for
aU
t>
also that i>c(0 )
c(+)

where

The

derivatives

of impulses.

Inductor

The inductor in Fig. 5.5a describes a dual relationship between voltage

and current when compared to a capacitor. The v-i relationships are

-'5
(5.4)

f ir)dr+iL(0-)

LJo-

where L is given in henrys. The initial current iL(0 ) can be regarded as

an independent current source, as shown in Fig. 5.56. As is true for the
voltage across the capacitor, the current through the inductor is similarly
continuous for all t, except in the case of impulse excitations.
When the network elements are interconnected, the resulting i-v equations are integrodifferential equations relating the excitation (voltage or
current sources) to the response (the voltages and currents of the elements).

There are basically two ways to write these network equations. The first
way is to use mesh equations and, the second, node equations.
Mesh equations are based upon Kirchhoff's voltage law. On the mesh
basis, we establish a fictitious set of loop currents with a given reference
direction, and write the equations for the sum of the voltages around the
loops. As the reader might recall from his previous studies, if the number
of branches in the network is B, and if the number of nodes is N, the
+ 1.*
number of independent loop equations for the network is B

See Stalling, op.

cit.

Network

anal/sis:

105

must, in addition, choose the mesh currents such that at least one mesh
current passes through every element in the network.

We

F.mpl 5.1.

In Fig.

a network is given with seven branches and five nodes.

1 = 3 independent mesh equations. The directions
i* / are chosen as indicated. We also note that the

5.6,

We therefore need 7 - 5 +
of the mesh currents

ilt

Vi(t)

Li

</W-

c,

Ci"
'1

biwQ

fCi'

R 2

vatQ

0^9-r Qhfi-i
FIG. 5.6

capacitors in the circuit have associated initial voltages. These initial voltages
are assigned reference polarities, as shown in the figure. Now we proceed to
write the

mesh equations.

Meshix

i(0

Mesh

c
I

if*

dr - J^t) dr

it :

vCl(0-)

Mesh

- *1 h(0 +
l

*C,(0-)

- vcfii-) = -

C*

^ J^r) dr+Li

di.

it :

-vJLO

+ vCt(0 -) =

^P

i/r) dr

^r

J\(t) dr+Ki tf

(5.5)

three unknowns, ilf it , and /, we can determine the branch

and the voltages across the elements. For example, if we were required
the branch currents i Cl and iCt through the capacitors, we would use the

After

we find the

currents

to find
following relationships

*<>*-'*-*
(5.6)
'c,

- '

Alternatively, if the voltage vj.t) in Fig. 5.6

v*t)

is

our objective, we see that

- tff)4

(5.7)

Network equations can also be written in terms of node equations,

which are based upon Kirchhoff's current law. If the number of nodes in

Network

106

anal/sis

and synthesis

Datum

FIG. 5.7

N, the number of independent node equations required

3
We can therefore
for the complete solution of the network is JV l.
node
voltages will be
the
network.
All
the
datum
node
in
select one

the network

is

datum node.
Consider the network in Fig. 5.7. Let us write a set of node equations
for the network with the datum node shown. Since the number of nodes
1
2 independent node equations.
in the network is
3, we need JV
These are written for nodes vx and vs , as given below.
positive with respect to this

Node

v^.

Node
j

B (t)

L(0~)

= G1 Vl (t) + j(* Vl (r) dr-jl* v t(r) dr

L Jo-

L Jo-

ty.

i(0-)

. --

!
*

v t (r)

L Jo-

dr

+ 7 Jf
L

' V 2(r) dr
o-

+C

^& + G

t> 2

()

(5.8)

dt

Further examples are given in the following sections.

5.3

INITIAL

AND

In this section,

FINAL CONDITIONS

we consider some methods for obtaining initial conditions

for circuit differential equations.

We also examine ways to obtain particular

networks with constant (d-c) or sinusoidal (a-c) excitations.

In the solution of network differential equations, the complementary
function is called the transient solution or free response. The particular
integral is known as the forced response. In the case of constant or periodic
oo is the steady-state or final
excitations, the forced response at t
integrals for

solution.

See Stalling, op.

cit.

Network

analysis:

107

There are two ways to obtain initial conditions at / = 0+ for a network:

through the differential equations describing the network, (b) through
knowledge of the physical behavior of the R, L, and C elements in the
(a)

network.
Initial

conditions for a capacitor

For a capacitor, the voltage-current relationship

i(r)dr

c(0+)
If

i(t)

at

= 0+

is

+ vc(0-)

(5.9)

Jo-

does not contain impulses or derivatives of impulses, c(0+)

If? the charge on the capacitor at t
0, the initial voltage is

o(0+)

vc<0-)

(5.10)

clude that

is no initial charge on the capacitor,

o(0+) = 0. We conwhen there is no stored energy on a capacitor, its equivalent

circuit at t

= 0+ is a short circuit.

When

there

This analogy is confirmed by examining

As a result of the conservation of

charge principle, an instantaneous change in voltage across a capacitor
implies instantaneous change in charge, which in turn means infinite
the physical behavior of the capacitor.

current through the capacitor.

Since

we

never encounter infinite current

in physical situations, the voltage across a capacitor cannot change

instantaneously. Therefore at

voltage source
initial

if

an

initial

= 0+, we can replace the capacitor by a

charge

exists,

or by a short

circuit if there is

no

charge.

Example 5.2. Consider the R-C network in Fig. 5.8a. The switch is closed at
t = 0, and we assume there is no initial charge in the capacitor. Let us find
the initial conditions i(0 +) and i'(0 +) for the differential equation of the circuit
Vu(.t)=Ri{t)

e',(t)/t

(5.11)

i(t)

>B

R-C network

R-C network,

'

>+))

T
m

(b)

>R

Equivalent circuit at <

FIG. 5A. (a)

A-

Equivalent circuit at /

- 0+

0+.

Network

108

The

and synthesis

analysis

equivalent circuit at

- 0+

is

given in Fig. 5.86, from which

we

obtain

>+>-5
To

obtain /'(u+),

we must

refer to the differential equation

V8(t) -Ri'(t)

At

(5.12)

i(0
(5.13)

(0+)
-Jt/'(0+) +'

= 0+ we have

(5.14)

We then obtain

i'(0+)-

-i(0+)
RC

(5.15)

Fig. 5.8a is
final condition, or steady-state solution, for the current in
obtained from our knowledge of d-c circuits. We know that for a d-c excitation,
is
a capacitor is an open circuit for d-c current. Thus the steady-state current

The

i(0
Initial

= '() -

conditions for an inductor

For an inductor, the voltage-current

1

relationship at

= 0+

is

iKr)dT+iL(0-)

(5.16)

If there is no
iL(0+) i(0-).
which corresponds to an open circuit at
_ o+. This analogy can also be obtained from the fact that the current
t
through an inductor cannot change instantaneously due to the conserva-

If v(t) does not contain impulses, then

initial current,

L(0+)

= 0,

tion of flux linkages.

S3 In Fig. 5.9a, the switch closes at t = 0. Let us find the

conditions i(0+) and i'(0+) for the differential equation
Example

L^+JI*)
L

o.c.

-o

9
(a)

FIG.

5.9. (a)

R-L network.

s=:

o-

i(o+)J

W
(*) Equivalent circuit at / *=

0+.

initial

Network

From

the equivalent circuit at

= 0+, shown in Fig.

5.9b,

analysis:

we

i(0+)=0

+)

/'(0

109

see that
(5.18)

We then refer to the differential equation to obtain <"(0+).

V-L i"(0 +) + R i(0 +)
Thus

(5.19)

V R
- - - - /(0 +)
(5.20)
as

The

V
L

steady-state solution for the circuit in Fig. 5.9a is obtained through the

knowledge that for a d-c source, an indicator

/(')

=/(>)

is

a short

circuit.

(5.21)

Example 5.4. In this example we consider the two-loop network of Fig. 5.10.
As in Examples 5.2 and 5.3, we use the equivalent circuit models at t +
and t = oo to obtain the initial conditions and steady-state solutions. At
f =
the switch closes. The equivalent circuits at t = 0+ and / = <x> are
shown in Fig. 5.1 la and b, respectively. The initial currents are
*i(0+)
ig(0+)

The

V
^~
*i

(5.22)

steady-state solutions are

) - j^r^ = ")

<5 - 23 >

Final conditions for sinusoidal excitations

When

a pure sinusoid, the steady-state currents and

same frequency as the
excitation. If the unknown is a voltage, for example, 1^(0. the steady-state
the excitation

is

voltages in the circuit are also sinusoids of the

solution

would take the form

i,(0

KOo)l

sin

K< - <K<*J\

(5.24)

and \V(jcoJ\ and ^(<u )

and phase of vl9(t). A similar expression would
hold if the unknown were a current.
To obtain the magnitude and phase, we follow standard procedures in
a-c circuit analysis. For example, consider the R-C circuit in Fig. 5.12.
The current generator is
( 5 -25)
W) - (Jo sJ n *) "(0

where

is

the frequency of the excitation,

represent the magnitude

Network

10

analysis

and synthesis
*2

VWV

h(0

-VW\r
^=:C

3
kM

FIG. 5.10

JVW\r

V!=

ww*

2(0+)

i(0+)J

Equivalent circuit at

O.C.

= 0+

WW-

-WWo
o.c.

Equivalent circuit at t

FIG.

5.1

1,

(a) Equivalent circuit at t

= 0+.

itfO

Y
FIG. 5.12

(6)

>

Equivalent circuit at

oo.

'

Network
form shown

If the steady-state voltage takes the

analysis:

III

in Eq. 5.24,

IHM>)I
|rOo)l

(5.26)

h
(G a

#o>

and
so that

We
5.4

vjLO

refer to this

^w

tan"

co

2/~*\W

*C*j

(G>+

problem in Section

(5.27)

"

sin (>o<

tan"

^)

(5.28)

5.5.

AND IMPULSE RESPONSE

STEP

As an

introduction to the topic of solution of network differential

equations, let us consider the important problem of obtaining the step and

impulse responses for any voltage or

As we shall see in

current in the network.

Chapter

the step and impulse responses

7,

are precise time-domain characterizations

of the network. The problem of obtaining

the step and impulse response
follows
energy,

is

stated as

Given a network with zero

we

initial

are required to solve for a

specified response (current or voltage)

F,e * 5 * 3

due

to a given excitation function u(t) or <3(0,

which either can be a current or a voltage source.

If the excitation

is

a step of voltage, the physical analogy is that of a switch closing at

time t
which connects a 1-v battery to a circuit. The physical
analogy of an impulse excitation is that of a very short (compared to the
time constants of the circuit) pulse with large amplitude.
The problems involved can best be illustrated by means of examples.
Consider the series R-C circuit in Fig. 5.13. The differential equation of

the circuit

is

KO =
We assume Vq(0) =
x'(t) for j'(0 in

0.

*)

= R 1(0 +
l(r) dr
7
J~
CJi>C
I

ii

Since Eq. 5.29 contains an integral,

(5.29)

we substitute

the equation, giving us

d(t)

= Rx'(t) + z(t)

(5.30)

Network

12

analysis and synthesis

Integrating both sides between

and

0+

gives

*>+>-The

(5.31)

characteristic equation is

H(p)

and with

little

effort

- Rp +

(5.32)

we have

_ L ,-t/RC n(0
*(0-

m-

so that

R^'RC

6"'*

(5.33)

"]

(5 ' 34)

We

thus arrive at the current impulse

in Fig. 5.14a.
as the result of an impulse voltage excitation. In the process
we have also obtained the step response x(i) in Fig. 5. 146 since, by definition,
the derivative of the step response is the impulse response. The reader

which

is

response

shown

i(t)

should check this result using a step excitation of voltage.

In the second example consider the parallel R-C circuit in Fig. 5.12
7 (/), where / is a constant.
driven by a step current source i(f )

m
'

aro

R*C
(*)

FIG. 5.14. (a) Impulse response of

R-C circuit.

(A)

Step response of R-C circuit.

Network

analysis:

113

FIG. 5.15. (a) Step response of parallel

R-C circuit.

(A)

Impulse response of parallel

Jt-C circuit.

Assuming zero

initial

conditions, the differential equation

Jo (0

from which we obtain the

Gt<0

steady-state value

C^p

(5.35)

characteristic equation

H(p)

The

is

of KO

Cp

+G

(5.36)

is

(5.37)

H(0)

m
/o

FIG. 5.1*. Pulse excitation.

Network

14

analysis

and synthesis

FIG.

5.17.

Thus the complete solution

r(0

From the

initial

to pulse excitation.

for the voltage step response

is

= (Ke-i/BC + 7oR) u(t)

(5.38)

= 0, we obtain K =
IRC u(t)
t>(0 = JoR(l - e-*
)

condition (0+)

I^R

so that
(5.39)

Differentiating Eq. 5.39 gives us the voltage impulse response

f(0

The step and impulse responses

= l* e -tlRC (0
are plotted

(5.40)

on

Figs. 5.15a

T)]

and

b.

Suppose

the excitation in Fig. 5.12 were a pulse

KQ = IMt) shown

in Fig. 5.16.

u(t

Then, according to the superposition and time-

invariance postulates of linear systems, the response

K0 =
which

5.5

is

V*[(l

shown

(5.41)

e-" BO) u(t)

- (1 -

e-

(i

would be

- T)IBC
)u(t

T)]

(5.42)

in Fig. 5.17.

SOLUTION OF NETWORK EQUATIONS

we

our knowledge of differential equations

There are two important points in
network analysis: the writing of network equations and the solution of
these same equations. Network equations can be written on a mesh, node,
or mixed basis. The choice between mesh and node equations depends
largely upon the unknown quantities for which we must solve. For
In

this section

will apply

to the analysis of linear networks.

Network
instance, if the

mesh

write

unknown

quantity

On

equations.

across a certain element, then

the choice

quite arbitrary.

is

a branch

is

the other hand,

if

current,

for example,

it is

we wish

node equations are

If,

analysis:

115

preferable to

to find a voltage

In many cases
to find the voltage v

better.

we wish

across a resistor R, we can either find v directly by node equations or find

the branch current through the resistor and then multiply by R.

Example
network
source

in Fig.

is e(t)

when

5.18,

the voltage

= 2e-- ufjt) and c(0-) = 0.

The

Solution.

*(/)

Find the current /(f) for the

5.5.

differential

= Rl(t) + ^ j

equation

i(r)dr

is

+ v^O-)

=fcC-if

(5.43)

Or, in terms of the numerical values,

we

have

2~" u{t)

= i(0 + 2

(t)

dr

(5.44)

Differentiating both sides of Eq. 5.44,

2(5(0

To

obtain the

the limits

initial

= 0

and

we

obtain

dm + 2i(t)
- e- u u(t) =
dt

condition i(0+),
t

FIG. 5.18

= 0+

we must

to give (0+)

(5.45)

integrate Eq. 5.45 between

= 2. From

the characteristic

equation

H(p)=p+2*0
we obtain

/<#)
If

(5.46)

the complementary function as

we assume the

particular integral to be iv(t)

A = The incomplete

= JCr

solution

_1

//(-0.5)

(5.47)

= Ae~

bt
,

then

we

obtain

(5.48)

is

= Ke~* - <r "

From the initial condition i(0+) = 2, we obtain the final solution,
Kt) = (f-* - f*--") (0
(0

(5.49)

(5.50)

As we already noted in Section 5.3, in the solution of network differential

is called the free response, whereas
a particular integral, and in the case of constant or
oo is the steady-state
periodic excitation, the forced response at /
solution. Note that the free response is a function of the network elements

equations, the complementary function

the forced response

is

Network

16

alone and

is

analysis

and synthesis

On

independent of excitation.

the other hand, the forced

response depends on both the network and the excitation.

positive elements,
It is significant that, for networks which have only
the free response

is

made up of only damped exponential and/or

sinusoids

characwith constant peak amplitudes. In other words, the roots of the

example,
For
parts.
real
zero
or
negative
have
all
teristic equation H(p)
0.
a
a j(o, then Re (pj
is a root of H(p) written as
x
if

= ^

/!

fact is intuitively reasonable because, if a

This
a response that

bounded excitation produces

exponentially increasing, then conservation of energy is

passive
not preserved. This is one of the most important properties of a
parts
real
whose
characteristic equation contains only roots
is

network. If a

network
are zero or negative, and if theyo> axis roots are simple, then the
unstable*
is
network
the
which it describes is said to be stable; otherwise,
will be discussed
Stability is an important property of passive networks and
in greater detail later.

Example

For the R-C network

5.6,

Eq. 5.25, find the voltage

p^O-) -

v(t)

in Fig. 5.19 with the excitation given

across the capacitor;

it is

given that v(0-)

by

0.

have already obtained the particular integral in Eq. 5.28.

on a
us find the complementary function. The differential equation

Solution.

Now

let

node

basis is

We

C -r + Gv = I

(5.51)

sin mtiKt)

at

from which we obtain the characteristic equation as

H(p)

-Cp+G

(5.52)

vJfl-KT* *

so that

and the incomplete

solution

(5-53)

is

- to-*) + . o)>c w sin {<* - tan"* J (/)

(G +
-g
^
obtain

we
t<0
)
0,
condition
From the initial
*/)

a>C\

.(o^-KO^-Jf-^r^fe^^l^'V)From the argand diagram in

Fig. 5.20

we see that,

Consequently,

This is not a formal definition of stability,

but

it

(5.54)

suffices for the

moment.

(555)

Network

analysis:

117

lm

A
A
o

C
1

Be

MG.SM
From the complementary function
the circuit is T - C\G - RC.

in Eq. 5.53,

we see that the time constant of

kinds of free responses

Next, let us examine an example of the different
relative values of the
on
depend
that
equation
of a second-order network
in Fig. 5.21 ; let us
network
the
given
are
we
network elements. Suppose
equation
differential
the
for
v^t)
find the free response
(5.58)
'

Differentiating

both sides of Eq.

VJf)

The

LJ*-

dt

S.S8,

we have

- C p"(t) + G '(0 + <0.

(5.59)

characteristic equation is then

H(p)

- Cp + Gp + 1 =

In factored form, H(p)

c(j>

+ 1 P + j^)

(5.60)

is

#(p)

= G(p-J>i)0>-/>i)

(5.61)

-)

nasai

118

Network

and synthesis

analysis
v(t)

+ K2

FIG. 5.22. Overdamped response.

AB

Pi

where

2C

P*

LCI

2l\c)

(5.62)

There are three different kinds of responses depending upon whether

real, zero,

case

B is

or imaginary.

Bis

1.

real, that is,

a,>
lc

(1
then the free response

is

vdt)

= Kie- U-B)t + K e- u+BU

(5.63)

which is a sum of damped exponentials. In this case, the response is said

to be overdamped. An example of an overdamped response is shown in
Fig. 5.22.

case

2.

5=

0, that is,

(-T-\Cl
LC
= p = A
At
vc(t) = (^ + K t)e-

then

p1

so that

When B

(5.64)

0,

the response

is critically

damped, as shown in Fig.

v(t)

o
FIG. 5.23. Critically damped response.

5.23.

Network

analysis:

119

FIG. 5.24. Underdamped response.

case

3.

Letting

B is imaginary,

this case, the

damped

is,

B = jfl, we have
c(0

In

that

response

is

LC

(*i sin

said to

fit

+ Kt cos

(5.65)

fit)

be underdamped, and

is

shown by the

oscillatory curve in Fig. 5.24.

Example

5.7.
In this example we discuss the solution of a set of simultaneous
network equations. As in the previous examples, we rely upon physical reasoning
rather than formal mathematical operations to obtain the initial currents and
voltages as well as the steady-state solutions. In the network of Fig. S.25, the
switch S is thrown from position 1 to position 2 at t = 0. It is known that
prior to t = 0, the circuit had been in steady state. We make the idealized
assumption that the switch closes instantaneously at / = 0. Our task is to

find i(0

and

/,(/)

The values of the batteries

and the element values are given as

after the switch position changes.

Vl and V% are Vx = 2 v, Vt =

3 v;

L=
C=
The mesh equations

for

ix{t)

Rt = 0.5 n
= 2.o n

lh,

jig

if,

and

it(i)

after

are

Vt =Li\{t)+Rl i1{i)-R1 i)
-vc(0-)= -/Mito +

(5.66)

cL ^T)dr+{R
k

+R^i^t)

(5.67)

Network

120

analysis

and synthesis

t.o^onnnr^
?2

v2 ^:

hit))

FIG. 5.25

Since Eq. 5.67 contains an integral,

we differentiate it to

give

= -Rt i\(0 + ^ i#) + (*i + **) i'#)

(5.68)

Using Eqs. 5.66 and 5.68 as our system of equations, we obtain the characteristic

equation

Lp +'*!

-Ri

H(p)-

-+(R1+ RJp

-RiP

<569>

- URx + *lp* + (| + *i*l)p +

Substituting the

dement values

H(p)

The

into H(p),

- 2.5p* + 4p +

1.5

we have

- 2.5(p + lXp+ 0.6)

(5.70)

free responses are then

(5.71)

The

steady-state solutions for the

mesh

currents are obtained at

considering the circuit from a d-c viewpoint. The inductor

circuit and the capacitor is an open circuit; thus we have

is

oo

by

then a short

(5.72)

Now let us determine the initial currents and voltages, which, incidentally, have
sources are not
the same values at / - 0- and t - 0+ because the voltage

Network
impulses. Before the switch is thrown at

analysis:

121

= 0, the circuit with Kx as the voltage

source was at steady state. Consequently,

(0-)-K1 2v

ro

0-)-^-4amp

(5.73)

0-)-0
We next find i\(0+) from Eq. 5.66 at / - 0+.
Vt ~Li\{0+) + 4*i<P+) - RiW+)
Substituting numerical values into Eq. 5.74,
'

From Eq.

5.68 at / <-

amp/sec

*!+*. /'i(0+)

values of

initial

we find

0+, we obtain similarly

*'>+)

With these

1(0+)

(5.74)

it(t)

and

i/t),

- 0.2 amp/sec

we can

(5.75)

quickly arrive at the final

solutions
/i(r)

i/f)

which are plotted in Fig.

- (0.5*-* - 2.5e- + 6) i<0

- ( -0.5tf- + 0.5*-) n(f)

5.26.

6.0

5.5

h(t)

5j0

45 4.0,

o.io'

0.08

\w

0.06
0.04

O02
r
(>

0.5

1.0

IS

2.0

2.5

FIG. 5.26

3.0

33

40 43

Network

122

and synthesis

analysis

ANALYSIS OF TRANSFORMERS

5.6

According to Faraday's law of induction, a current ix flowing in a coil L x

induce a current it in a closed loop containing a second coil Lt The

may

for

inducing the

sufficient

conditions

current

are: (a) part of the flux

it

Ox

in

Lx must be coupled magnetically

coil L a (b) the flux O x must be

the coil
to the

changing with time.

In

this section

we

containing a device

will analyze circuits

made up of two
known as a

magnetically coupled coils

transformer. In Fig. 5.27, the schematic

of a transformer is given. The Lt side

of the transformer is usually referred to
as the primary coil and the L t side as the secondary coil. The only distinction
between primary and secondary is that the energy source is generally at
the primary side.
The transformer in Fig. 5.27 is described mathematically by the
FIG. 5.27. Transformer.

equations

VO-L^ + M^
at

at

(5.77)
, x

du

di t

at

at

is the mutual inductance associated with the

where
and is related to Lx and L2 by the relationship

flux linking

Lt to Lt

M = k4IT
x

The constant
bounded by the

K in
limits

Eq. 5.78

(5.78)

called the coefficient of coupling.

1. If \K\
1, then all of the flux

is

<, \K\ <,

It is

Ox in

Lj is linked magnetically to L % In this case, the transformer is a unitycoupled transformer. If

0, the coils L x and L s may be regarded as
two separate coils having no effect upon one another.
coil

K=

For

circuits

with transformers,

for the mutually induced voltages

we must

establish reference polarities

dijdt.

Usually, the references are

on the input and output leads of a transformer,

shown by the dots on the schematic in Fig. 5.28. The reference dots are

given by small dots painted

as

placed at the time of manufacture according to the procedure outlined

here.
voltage source v is connected to the primary Lx side of the
voltmeter is attached on the secondtransformer, as shown in Fig. 5.28.
terminal
is assigned the dot reference to
the
side,
ary. At the primary

Network

analysis:

123

Voltmeter

ttransformeri

An experiment

FIG. 5.28.

to determine dot references.

which we connect the positive lead of the voltage source. The dot reference
is placed on the secondary terminal at which the
voltmeter indicates a
positive voltage. In terms of the primary current i the positive
voltage at
lt
the secondary dot is due to the current i flowing into the dot
on the
t
primary side. Since the positive voltage at the secondary dot corresponds
to the current

we can think of the dot references

both currents are flowing into the dots or away
from the dots, then the sign of the mutual voltage term Mdijdt is positive.
When one current flows into a dot and the other away from the second dot,
the sign of Mdijdt is negative.
/,

flowing into that dot,

in the following way.

If

and

linkages of Ly

and

it

If

number of turns of coils L^ and L% then the flux

and iV.O,, respectively. If both ix
dots, the sum of flux linkages of the transformer is

are the

and L, are given by

flow into the

N&

2 $ linkages = N&i + N^
however, one of the currents, for example,
other ig flows out of the other dot, then
If,

ilt

(5.79)

flows into a dot, and the

2 $ linkages = N& - N Q
t

(5.8O)

An

important rule governing the behavior of a transformer is that the

sum of flux linkages is continuous with time.
The differential equations for the transformer in Fig. 5.29 are

V u(t) = Lt i\(t) + Rt i&) +

Q

i\(t)

R* i8(0

FIG. 5.29

i'tf)

+ L, I'M

(581)

Network

124

analysis

and synthesis

Integrating this set of equations between

= 0

and

= 0+

results in

the determinant

- /i(o-)] MM>+) - o-)] _

"Q
M\ii(f>+) - ix(0-)] LJMP+) - '.(0-)]

Za&(0+)

By

(I^L,
If
t

M)[i1(0+)

K<

obtain

*1 (0-)][i,(0+)

/,(0-)]

>

M*, that

is,

fl

(5.83)

1,

(0+)

*(0+)

and

K<\
*<1

- /,((>-),
-

ig(0-),

For the transformer circuit in Fig. 5.29,

Rt -= 3Q, R, 80, and Jlf - 1A. The excitation is V - //0-) - 0.
f (/) and i#), assuming that /x(0-)
irph

5.8.

Solution.

The

differential

6(0

The

then the currents must be continuous at

equal to zero. Thus
in order for the determinant in Eq. 5.82 to be

L^Lx

we

evaluating this determinant,

(5.82)

(5.84)

(5.85)

Lt 6ii(0-

1A,

2h,

Let us find

equations for the circuit are

- i'i(0 + 3i\(0 + /',(<)

- iVO + 2i'&) + 8i,(0

characteristic equation is given

(5.86)

by the determinant

Wj

+8
= 0> + 3)(2p + 8)-/>
= (^+14/ + 24)
-

(/>

2^

+ 2)(p +

(5.87)

12)

Thus the complementary functions are

= Kx<r + A^e-"'
= K*r + JV"m
Ie (0

ile{t)

(5.88)

obtain the particular integral, or steady-state solution,

physical reasoning to arrive at

To

V_

hJt

Oamp

we

rely

upon

amp
(5.89)

Network

analysis:

125

Since the excitation does not contain an impulse (and since I^L,
jt 0, as we shall see later), we can assume that
d(0+) and ^(0+) are

Af1

also zero.

Then using Eq.

5.86

we can find i\(0+) and i'J0+).

+ f,(0+)
o -r<p+) + 2i',(o+)
1(0+) - 12 and i'0+) - 6.
6'"i(0+)

Solving,
ditions,

(5.90)
'

we find i'
With
we obtain the final solutions of ix(t) and %(/)

these initial con-

(5 ' 91)

0 - (-Ir* + *e-0 (0
Suppose, now, I^L, - M, that
K - then i,(0 and (/) need not be
continuous at 0. In fact, we will* show that the currents are discontinuous at >
for a unity-coupled transformer. Assuming that K = 1,
is,

1,

consider the

mesh equation
*t 'i(0+)

for the secondary at i

= -M i'^O-H) - L, .',(0+)

(5.92)

--?[!* i'!(0+) + M i',(0+)]

(5.93)

The mesh equation of the primary

V - Rt

it

(0+)

[L,.

= 0+

i\(0+)

side then

becomes

+ M iV0+)) = Rr h(Q+) -^R, iJ0+)

(5.94)

We

need an additional equation to solve for ^(0+) and /g(0+). This

provided by the equation

iPi(0+)

it

+ MM>+) - i^O-)] =
5.82. Since x (0-) = i^0~) = 0,

(0-)}

which we obtained from Eq.

is

(5.95)

we

solve

Eqs. 5.94 and 5.95 directly to give

^3

1^(0+)

R\JL%

wo+)

=RL
1

Consider the following example.

the element values are

i=4h,

At-8 0,
M = 2h,

+ R^Li

(5.96)

+ R^

For the transformer

L,= lh
R9 =*3Q

K=10v

in Fig. 5.29

Network

126

anal/sis

and synthesis

Assuming that the circuit is

t

= 0, let us find h(t) and

at steady state before the switch

The

/2 (0-

is

differential equations written

closed at

on mesh

basis are

10-4^
+ 8^ + 2^
dt
dt
(5.97)

dt

The

characteristic equation

dt

is

4(p

2p

2)

H{p)
2p-

- 20p +

H(p)

which yields

(p

24

(5.98)

3)

= 20(p + |)

(5.99)

so that the complementary functions are

(5.100)

The

we

particular integrals that

(r)

obtain by inspection are

= 1 = 12 =

(5.101)

hit)

The

initial

conditions are

*i(0+)

i*(0+)

We then find Kx =

VL,

=
RtLi

+ R^

-0.75 and

20

(5.102)

-VM

-!2-a5
=

K = - 1.0 so that
t

= (-O.lSe-11 +
W) = -er^u{t)
*

ijff)

We

see that as

state value

of

approaches

1.25.

-1.0

infinity

it(t)

1.25) u(t)

- 0, while

(5.103)

it (t)

goes

its

steady-

Network

analysis:

127

Problems
5.1

Write the mesh equations for the network shown.

nnnp

nnnp

Cl <0->

Pi

*^\{=

'l(0-)

2<0-)

*~-,

Ci

*1

QWb(

te1

(0-)=J;C2

PROB.
5.2

Write a

set

J8 2

5.1

of node equations to solve for the voltage vjt) shown in the

figure.

W
ii

wm{(T)

SJlo

Ci

*2>

= = G>

2(0

c <0-)

PROB.

5.2

53 The network shown has reached steady state before the switch S is
opened at t = 0. Determine the initial conditions for the currents
1-fy) and
4(0 and their derivatives.

R
o^f o

vW-

^WV

2WJ

1
PROB. 5.3
5.4

The network shown has reached steady

from a to

b.

derivatives.

state before the switch

moves

Determine the initial conditions for iL(t) and vdt) and their
Determine also the final values for iL(t) and vc(t).

first

128

Network

analysis and synthesis

1
10 v

lh_

vc (t)

*t

5v-

==

If

:io

T
PROB.

5.4

5.5 The network shown has reached steady state before the switch moves
from a to b. Determine the initial conditions for the voltages vx(,t) and vt(t)
and their first derivatives. Determine also the final values for v^t) and vjj).

t0

vi(t)

J:-

*i

=C

R2 *

v 2 (t)

i
PROB.

5.6

For the network shown ii{0 )

0- < / < .
(/);

(a)

Find

(A)

Show

(c)

5.5

= 0.

that v(t) approaches an impulse as

Find the strength (area) of the impulse.

- 0.

ItfOQ)

PROB. 5.6

5.7 For the network shown,

and sketch Jbto; 0- < r < .

(r)

d(t)

- tr*u(t)

and t<0-)

= 4.

Find


Network
10
I

v/W-

analysis:

129

o
m
\k*t)

1Q

W(J)

PROB.

5.8

5.7

For the network shown, before the switch moves from a to

Find the current /(f).

b, steady-

state conditions prevailed.

6<?

PROB.

5.9

For the

been in steady

5.8

shown, switch S is opened

Find /(f); 0 <f < .

circuit

state.

at

after the circuit

had

>Tnnp

/^
PROB.

U\

5.9

5.10 Find the current /(f) in the network shown when the voltage source is a
unit impulse. Discuss the three different kinds of impulse response waveforms
possible depending upon the relative values of R, L, and C. All initial conditions
are zero at/ 0.

130

Network

anal/sis

and synthesis

nnRp

Qm

*>J

PROB.

5.10

5.11 An R-C differentiator circuit is shown in the figure. Find the requirements for the R-C time constant, such that the output voltage v^t) is approxi-

mately the derivative of the input voltage.

uoW

PROB. 5.

5.12 An R-C integrator circuit is shown in the figure. Find the requirements
for the time constant such that the output t> (f) is approximately the integral of
the input voltage.

S
A/V

O"

ci

")
PROB.

5.13 Find the free response for

source; (b) a voltage source.

i(f)

vo(t)

5.12

in the figure

shown

for (a)

a current

Network

analysis:

131

Source

PROB. 5.13
At t = 0, the switch goes from position 1 to 2. Find /(/), given
= e~* sin It. Assume the circuit had been in steady state for t < 0.

5.14
e(t)

PROB.

5.15
initial

that

5.14

For the circuit shown, switch S closes at t = 0. Solve for l(t) when the
conditions arc zero and the voltage source is e(t) = sin (cat + 0). What

should be in terms of R, C, and

term is zero?

so that the coefficient of the free response

-VNAr
I"

^)
PROB.

5.15

5.16 For the circuit shown, the switch S moves from a to b

and sketch v^t) for - < / < . The circuit is in a steady

at

= 0.

state at

Find
f

= 0.

Network

132

analysis

and synthesis

MA

->

Xs

1
PROB.

5.17

For the

circuit

shown,

i(r)

5.16

- 4r (0-

Find

i<f);

- </<

JQ^

lf=t=

PROB.

ih

5.17

in Prob. 5.4.
5.18 Find the complete solution for i L(t) and vdO
Find
t - 0.
5.19 For the circuit shown, the switch is closed at
energy.
initial
zero
oo.
Assume
<t <
ijj) for

10Q

irff)

ana

lh

10Q

PROB. 5.19
For the transformer circuit shown, the switch
and i2(f). Assume zero initial energy.
S

5.20
ijft)

5i>.

Af=lh
PROB. 5.20

closes at

t = 0.

Find

Network

Q>

30

PROB.

lh<3 1hg>lh

analysis:

133

60^02(0

5.21

switch S opens at
5.21 For the transformer circuit shown, the
. Assume zero initial energy.
/
the voltage far) for

- < <

= 0.

Find

chapter

The

Laplace transform

THE PHILOSOPHY OF TRANSFORM METHODS

6.1

In Chapters 4 and
ential equations.

discussed classical methods for solving differsolutions were obtained directly in the time domain

5,

The

we

of solving the differential equation, we deal with

time
at
every step. In this chapter, we will use Laplace
functions of
transforms to transform the differential equation to the frequency domain,
where the independent variable is complex frequency s. It will be shown
that differentiation and integration in the time domain are transformed
into algebraic operations. Thus, the solution is obtained by simple
since, in the process

algebraic operations in the frequency domain.

a striking analogy between the use of transform methods to

solve differential equations and the use of logarithms for arithmetic
operations. Suppose we are given two real numbers a and b. Let us find

There

is

the product

c=axb

(6.1)

by means of logarithms. Since the logarithm of a product is the sum of the

logarithms of the individual terms,
log

so that
If a

and b were two

we have

C = log a X b = log a + log b

C = logr1 (log a + log b)

six-digit

(6.2)

(6.3)

numbers, the use of logarithms would probably

facilitate the calculations, because logarithms transform multiplication

into addition.
is the use of transform methods to solve integroConsider the linear differential equation

An analogous process
differential equations.

yW))

-/()

134

(6.4)

The Laplace transform

135

Algebraic

Differential

equation^

equation

Transform

FIG.

6.1.

X(s)
^

x(t)

Solve

Invert

Philosophy of transform methods.

the forcing function, x(t) is the unknown, and y(a<0) is the

differential equation. Let us denote the transformation process by 7X-),

where /(0

and
Eq.

is

be the frequency variable.

we have

When we

let s

6.4,

= W)]

(6.5)

domain functions are given by

capital letters, let us write

7W*(0)1

Since frequency

Eq-6.5as

Y(X(s),s)

where X(s)

transform both sides of

TftOL

*W -

= F(s)

T[f(t)\,

and

(6.6)

Y(X(s), s)

is

an algebraic

equation in s. The essence of the transformation process is that differential

equations in time are changed into algebraic equations in frequency.

We

can then solve Eq. 6.6 algebraically to

obtain X(s). As a final step, we perform
an inverse transformation to obtain

*(0

r-

(6.7)

[*(*)]

Differential

equation

EM

System

m
RM

function

In effecting the transition between the

FIG. 6.2. Linear system.
time and frequency domains, a table of
very
transform pairs MO, X(s)} can be
diagram outlining the use of transform methods is given
helpful.
in Fig. 6.1. Figure 6.2 shows the relationship between excitation and
response in both the time and frequency domains.

THE LAPLACE TRANSFORM

6.2

The Laplace transform of a function of time/(/)

CLf (01

= Hs) =

{"/(&-" dt

is

defined as

(6.8)

no-

where s

is

the complex frequency variable

s

a +ja>

(6.9)

This definition of the Laplace transform is different from the definition

given in most standard texts, 1 in that the lower limit of integration is
1

M.

New

E.

Van Valkenburg, Network Analysis, 2nd Ed. Prentice-Hall, Englewood Cliffs,

Jersey, 1964.

Network

136
/

= 0

instead of

that/(t)

and synthesis

= 0+. We

thus take into account the possibility

may be an impulse or one of its higher derivatives.

C[d(0]

analysis

for the

It is clear that

0+ definition, whereas for the 0- definition, C[($(0] =

In the case when no impulses or higher derivatives of impulses are involved,

Therefore, all of the
it was shown in Chapter 4 that/(0-) =/(0+).
treatment
of Laplace trans"strong results" resulting from a rigorous
lower
limit
also
apply for the 0
a
0+ as
forms* obtained by using t

definition.

In order for a function to possess a Laplace transform,

it

must obey the

condition

\f{t)\e-"dt<

oo

(6.10)

Jo-

Note that, for a function to have a Fourier

must obey the condition

for a real, positive a.

form,

it

f
/

|/(0I

dt<

oo

trans-

(6.11)

As a result, a ramp function or a step function will not possess a Fourier

transform, 3 but will have a Laplace transform because of the added
t%
convergence factor er a *. However, the function e will not even have a
Laplace transform. In transient problems, the Laplace transform is
preferred to the Fourier transform, not only because a larger class of

waveforms have Laplace transforms but also because the Laplace transform takes directly into account initial conditions at t = 0 because of
the lower limit of integration in the Laplace transform. In contrast, the
Fourier transform has limits of integration ( oo, oo), and, in order to take
into account initial conditions

function must take a form

due to a switch closing

as/(0 u(t)

+ /(0-)

(3(0,

at

= 0, the forcing

where /(0-) repre-

sents the initial condition.

The

inverse transform C-lrjFT(.0]

/(0

is

= -M

FWds

(6.12)

Alt} Jniaa

a real positive quantity that is greater than the a convergence

factor in Eq. 6.10. Note that the inverse transform, as defined, involves a

where at

is

D. V. Widder, The Laplace Transform, Princeton University Press, Princeton, 1941.

These functions do not possess Fourier transforms in the strict sense, but many
possess a generalized Fourier transform containing impulses in frequency; see M. J.
Lighthill, Fourier Analysis and Generalized Functions, Cambridge University Press, New
*
*

York, 1959.

The Laplace transform

137

complex integration known as a contour integration.1 Since it is beyond

the intended scope of this book to cover contour integration, we will use a
partial fraction expansion procedure to obtain the inverse transform.

by

find inverse transforms

recognition,

we must remember

To

certain basic

transform pairs and also use a table of Laplace transforms. Two of the
most basic transform pairs are discussed here. Consider first the transform

of a unit step function

u(t).

Example 6.1. /(/)-(/).

(6,13)

Next,

let

us find the transform of an exponential function of time.

Example 6.2. /(/)

~ e<(01

F(j)-J
With

these

e#e-dt=s

(6.14)

two transform pairs, and with the use of the properties of Laplace
we discuss in Section 6.3, we can build up an extensive table

transforms, which

of transform

6.3

pairs.

PROPERTIES OF LAPLACE TRANSFORMS

In this section we will discuss a number of important properties of

Laplace transforms. Using these properties we will build up a table of
transforms. To facilitate this task, each property is illustrated by considering the transforms of important signal waveforms. First let us
discuss the linearity property.

Linearity
finite sum of time functions
forms of the individual functions, that is

The transform of a

t[| /,]=

is

the

sum of the trans-

2 OK')]

(615)

This property follows readily from the definition of the Laplace transform.

Example 63. /(/)

= sin <ot.

Expanding

fit)

For a

by Euler's

identity,

we have

= ^.(e** - r+*)

Goldman, Transformation Calculus and

Englewood Cliffs, New Jersey, 1949, Chapter 7.

lucid treatment, see S.

Transients, Prentice-Hall,

sin cat

(6.16)

Electrical

38

Network

anal/sis

and synthesis

The Laplace transform of /(f)

cisoidal e iat terms. Thus

is

the

sum of

1/1

=-

C[sin cat]

\
1

yto
:

2/\*

the transforms of the individual

+.//

=
**

(6.17)

>*

Real differentiation
Given that [/(f)] = *(*), then

Eflwhere /(0-)

is

By

Proof.

sF(s)-/(0-)

the value of/(f) at

(6.18)

= 0-.

definition,

[A0]=J*V'7'(0<*f
Integrating Eq. 6.19

by

parts,

[/'(')]

(6.19)

we have

= -7(0

/(Oe-'df

(6.20)

JoSince e~" -
[/"(')]

as

-*

oo,

and because the

= ^) we have
t[f'(t)]

Similarly,

we can show

|"7(0"l

n
s

f(s)

integral

on the right-hand

= sF(s)-f(0-)

side

is

(6.21)

for the nth derivative

-i
s

/(0 _)

_ S "-V(0-)

- u (0-)
(6.22)

where f

"-"

is

the (

that differentiating

by

l)st derivative

in the time

of /(f) at

domain

is

= 0-. We

thus see

equivalent to multiplying

complex frequency domain. In addition, the initial conditions

by the terms / (i) (0 ). It is this property that
transforms differential equations in the time domain to algebraic equations
in the frequency domain.

by

s in the

are taken into account

Example

6.4a.

/(f)

= sin cot.

Let us find

By

j sin

cot)

(6.23)

cgsin<of]=,(^)

(6.24)

C[cos

cot]

the real differentiation property,

we have

The

i(_^ = _I 5

c[cosa,,]=

sothat

In this example, note that sin <o(0)

Example

6.4b.

unit impulse.

/(/) = (/). Let

We know that

Laplace transform

139

(6.25)

0.

us find [/'(')], which

the transform of the

is

=-

Ct(0]

(6.26)

C[*0]

Then
since

= *(j) =

(6.27)

= 0.

(0 )

Real integration
= F(s)> then the Laplace transform of the integral of /(f)

If [/(*)]
flfa)

divided by

that

s,

c[/(r)dr]=^
By

Proof.

c|"

by

LX

parts,

/(t)

-.C

[-

/(T) dT
]

d'

(6,29)

obtain

=-

/(t) dr

s Jo-

J
as

e_rt

/(t) dr]

we

rfrl

LJoSince e~" -*

(6.28)

definition,

c
Integrating

is

is,

oo,

and

"

+-

o-

e-'/(f) dt

(6.30)

s Jo-

since

=0

/(t)<*V

Jo-

(6.31)

It-o-

we then have

c[/(r)dr]=^
Example 6.5.

We know that
Since

Let us find the transform of the unit ramp function, p(t)

h(t)

dr

C[(0]

(6.32)

" t u(t).
(6.33)

p(f)

=-

(6.34)

then

C[p</)]

-^ . _
[(')]

(6.35)

Network

140

and synthesis

analysis

Differentiation by
Differentiation

by

multiplication

by

s in the

in the

complex frequency domain corresponds to

domain, that

W)] =
From the

Proof.

is,

-^

(6-36)

as

of the Laplace transform, we see that

definition

1 = f7(0 e-* dt - - f " tf(t)e-* dt Example 6.6.

Given /(O

-Z[t f(t)]

= g-*\ whose transform is

mrrz
let

us find

C[te~"']-

the preceding theorem,

By

638>

we find that

"-i--s(rh)-<rb
Similarly,

we can show

that

(640)

"a-^-crrb* 5
where n

is

a positive

Complex
By

integer.

translation

the complex translation property,

/=X5

where a

is

Proof.

(6.37)

Jo-

ds

Jo-

ds

if

F(s) <= [/(/)], then

- a) = C[e7(0]

< 6 - 41 )

a complex number.

By

definition,

1*7(0]
Example 6.7.

-J"

^/(Oe"" dt

Given /(t)

J"

= sin </, find

.

e"

(M)

7(0 dt

[-' sin </].

qsinorfl-^^

= F(s -

a)

(6.42)

Since

643>

<

by the preceding theorem, we find that

Similarly,

we can show that

c^coso,.]-

^;^;^

(-45)

The Laplace transform

141

Real translation (shifting theorem)

Here we consider the very important concept of the transform of a
P(s), then the transform
shifted or delayed function of time. If C[/(/)]
of the function delayed by time a is

[/('

Proof.

By

(6.46)

= f V*7(r - a) dt

(6.47)

a)]

definition,

C[/(r

- a) u{t -

new dummy

Introducing a

C[/(t) m(t)]

= e-** F(s)

- a) u(t -

f
Jo-

a)]

variable t

= a, we have
t

= g ['f(r)e-

e-lT^f(r) dr

dr

= <f~ F(s)

(6.48)

Jo-

in Eq. 6.48/(t)u(t) is the shifted or delayed time function; therefore

the theorem is proved.
It is important to recognize that the term e~"' is a time-delay operator.
If

we

are given the function

e~"

G(s),

and are required to

C-i[r~G(,)]-

we can

discard er*' for the

moment,

find

ft (0,

find the inverse transform

E-HG(*)]-('X
and then take

into account the time delay

by

setting

g(t-a)u(t-a)=g1(t)

(6.49)

Example 6.8a. Given the square pulse /(f) in Fig. 6.3, let us first find its transform F(s). Then let us determine the inverse transform of the square of F(s),
i.e., let

us find

Mt) =
Solution.

The square pulse

is

/(/)
Its

Laplace transform

is

C-i[F(5)]

(6.50)

given in terms of step functions as

- (/) -uit-a)

(6.51)

then
fit)

F(s)=\(l-e~)
Squaring F(s),

F\s)

we

(6.52)

obtain

- -5 (1 - 2e- + e"*")

(6.53)

To find the inverse transform of F\s), we

need only to determine the inverse transform

FIG.

4.3.

Square pulse.

Network

142

analysis

and synthesis

of the term with zero delay, which

is

&]-

C-4-J -fN(<)
Then

C-[F*(*)]

/(/)-

2(*

a)u(t

(6.54)

- a) +(t - 2a) (/ - 2a)

(6.55)

so that the resulting waveform is shown in Fig. 6.4. From this example, we see
that the square of a transformed function does not correspond to the square
of its inverse transform.

AW
f(t)

K6

K2
Ki

Xb

it

Kq,,

I
a
FIG.

FIG.

Triangular pulse.

6.4.

Example

6.8b. In Fig.
of a train of impulses

fit)

6.5, the

6.5.

Impulse

output of an ideal sampler

= *o <K0 + *i <K' ~ Tt) +

The Laplace transform of this impulse

C[/(/)l

3Ti 4Ti 5Ti 6Ti

2T!

Ti

2a

"

is

7Ti

train.

shown.

It

consists

+ * *' - ri)

656>

train is

- *o + *!*-* + ^-2,Tl +

In dealing with sampled signals, the substitution z =

represent the transform of the impulse train as

+ *e-* r
e ,r

is

often used.

(6.57)

Then

we can

Ai1

W)]-*. + T

A.

+^ +

A
+?

(6.58)

in Eq. 6.58 is called the z-transform of /(/). This transform

widely used in connection with sampled-data control systems.

The transform

PERIODIC

is

WAVEFORMS

The Laplace transform of a periodic waveform can be obtained in two

ways (a) through summation of an infinite series as illustrated in Example
6.8d, and (*) through the formula derived below.
:

[/(*)]

-}

f(tyr"dt

+Jt

f{t)e-'*dt

+
(6.59)

flufUT
f(t)e-*dt
InT

The Laplace transform

Since /(0

is

periodic, Eq. 6.59 reduces to

[/]

T
dt
\ f{i)e-'*dt
Jo-

e- T [
+ e-'

e-nT f*f(t)e-* dt

= (1 +

e-T

l-e-

+ e-aT +

(6.60)

T
\ f{t)e-*dt
Jo-

T
~
J
T Jo- f(f)e-'*dt
\

Given the periodic pulse

6.8c.

determine

f(t)e-*dt

Joo-

Jo-

Example

143

train in Fig. 6.6 let us use Eq. 6.60 to

Laplace transform.

its

dt

-1
(6.61)

1 1

- ~"
- e-' T

m
1

a
FIG.

Example

6.8d.

Fig. 6.6 using

T+a

2T+a

2T

6.6. Periodic pulse train.

In this example we calculate the Laplace transform of /(f) in

summation of an infinite series. The periodic pulse train can be

represented as
/(/)
Its

iKt)

- iKt - a) + u(t - T) - u[t - (T + a)]

+ u(t - IT) - u[t - (2T + a)] +

Laplace transform
F(s)

(1

{pM)

is

- *-" + e~tT - e-<T+) +

.)

(6.63)

=s

[1

- e~" +

e~,T(l

- e-) +

*-2 T(l

- -") +

Network

144

which

analysis

and synthesis

simplifies further to give

F(s)

(1

- -^"Xl + e- T + e~UT +)

(6.64)

We then see that F(s) can be given in the closed form

1 1 - e-
Other periodic pulse trains also can be given in closed form. The reader
end of this chapter.

is

referred to the problems at the

At

the end of the chapter

is

a table of Laplace transforms. Most of

the entries are obtained through simple applications of the properties just
discussed. It is important to keep those properties in mind, because many

transform pairs that are not given in the table can be obtained by using
For example, let us find the inverse transform of

these properties.

F(s)

=,

j_

(s

ay

(6-66)

,
car

Since the s in the numerator implies differentiation in the time domain,

we

can write

"

F(s)-a
(s

Using the differentiation property,

we

ay

(6.67)

co'

obtain

dt

Note that e~"

6.4

sin

mt at

(6.68)

= (r sin oat +
= 0 is zero.

to

cos oot)e~*

USES OF LAPLACE TRANSFORMS

Evaluation of definite integrals

The Laplace transform

integrals.

is

often useful in the evaluation of definite

occurs in the evaluation of

An obvious example

-/:c-**sin5rdr
If we replace tr %%

of sin

5/,

which

(6.69)

by er**, the integral /then becomes the Laplace transform

is

Z[S ux5t]

= -rj-s + 25

(6.70)

The Laplace transform

Replacing J by

2,

we have
I

TTs

Perhaps a more subtle example

/
First,

we note

that

eriW

is

is

(6 - 71)

the evaluation of

+
J

,W dt

(6.72)

an even function; therefore

= 2jVC-2'df

From

145

the table of Laplace transforms

we

(6.73)

see that

and the transform of

= f tV**Jo-

(6.75)

vm -^tv

(<76)

/(0

Later

we

consists

will see that the partial-fraction expansion of Z[f(t)] in Eq. 6.76

of three terms for the multiple root (s + If, as given by

pr//A1
ELKO]

= 0.25 "

0.25

Taking the inverse transform of Eq.

f{t)

Now
t

1,

^i - j^
0.5

6.77,

= 0.25(1 - e- -

we

2te-**

.-_
(6.77)

obtain

- 2rtr-)

observe that the definite integral in Eq. 6.73

that

11(f)

(6.78)

is

equal to 2/(f) at

+ 0.5 = 0.162

(6.79)

is,

= 2/(1) =

-2.5<r*

Solution of integrodifferential equations

In Section 6.3 we said that the real differentiation and real integration
properties of the Laplace transform change differential equations in time
to algebraic equations in frequency. Let us consider some examples
using Laplace transforms in solving differential equations.
Example

6.9.

Let us solve the

differential

equation

+ 3x'(t) + Irtf) - 4e*

given the initial conditions a<0-) = 1, x'(Q) = 1.
*"(/)

(6.80)

Network

146

Solution.

analysis

and synthesis

We first proceed by taking the Laplace transform of the differential

equation, which then becomes

[i

X(s)

-sx(0-)- x'(0 -)] +

3[s X(s)

- x(0 -)] + 2X(s) =

Substituting the initial conditions into Eq. 6.81

(**
(s*

We then obtain

3s

+ 2) X(s)

and

simplifying,

we have

+5+2

(6.82)

X(s) explicitly as

-jr=mrm+%
To

(6.81)

find the inverse transform x(t)

= f'^W],

(6 - 83)

we expand

X(s) into partial

fractions

Solving for X_i,

The

A^i,

and

Kt algebraically, we obtain
K_i = JTi = -1 Kt = |

final solution is thf inverse

a<0

transform of X(s) or

fe*

- e-' + fe-

(6.85)

In order to compare the Laplace transform method to the classical method of

is referred to the example in Chapter 4,

solving differential equations, the reader

Eq. 4.64, where the differential equation in Eq. 6.80

Example

6.10.

Given the
2x'(f)

set of

is

solved classically.

simultaneous differential equations

4a<r)

+ y'(t) + 7y(0 = 5u(t)

+ y'(t) + 3jK0 = 56X0

(6.86)

+ x(t)
with the initial conditions x(0 -) = y(0 -) = 0, let us find x(j) and y(t).
x'(t)

Solution.

Transforming the
2(s

(s

set

of equations,

+ 2) X(s) + (* + 7)
+

1)

X(s)

+ (* +

Solving for X(s) and Y(s) simultaneously,

X(s)

we

obtain

Y(s)

--

3) Y(s)

(6.87)

we have

+ 5^

(5/*)A u

A
(6.88)

Y(s)

(5/5)Au

+ 5A*

The Laplace transform

where
zyth

is

the determinant of the set of equations in Eq. 6.87, and

More explicitly,

cofactor of A.

X(s)

-5i

in partial fractions,

- 30s +

the

15

(689)

we have

Kx

+ Kz

K-s

^> = 7 + ?T27T5
Multiplying both sides of Eq. 6.90 by s and letting *

K1

is

is

~-*FT2r+sr
Expanding X(s)

A tf

147

<690)

= 0, we find

=sX(s)\,_ =3.

Kt and Ks are then obtained from the equation

-8s

- 36

* "-7+5 + 5

(691)

A further simplification occurs by completing the square of the denominator

of X(s), that

is,

s*

As a

result of

+ 2s +

(s

1)*

+4

(6.92)

Eq. 6.92, we can rewrite X(s) as

3

*<'>-"

8(i
fr

so that the inverse transform

(/)

In similar fashion,

Wrt
y(s)

The

we

_
=

(3

+ 1) + 14(2)
+ iy+gy

(693)

is

8c-' cos 2/

14e-' sin It) (/)

(6.94)

obtain Y(s) as

"* + 17
1
+
~s 7T27T5 - -; +
*

+ l)+3(2)
+ i)* + (2)*

llfr

(,

(6,95)

inverse transform is then seen to be

y (j)

= ( _ i +1 i e-* cos 2f +

3<r* sin It) u(t)

(6.96)

is also solved by classical methods in Chapter 4, Eq. 4.163.

note one sharp point of contrast. While we had to find the initial
conditions at t
0+ in order to solve the differential equations directly
in the time domain, the Laplace transform method works directly with

This example

We

= 0

In addition, we obtain both the complementary function and the particular integral in a single operation when we
use Laplace transforms. These are the reasons why the Laplace transform
method is so effective in the solution of differential equations.
the

initial

conditions at

Network

148

analysis

and synthesis

PARTIAL-FRACTION EXPANSIONS

&5

As we have seen, the ease with which we use transform methods depends
upon how quickly we are able to obtain the partial-fraction expansion of a
given transform function. In this section we will elaborate on some simple
and effective methods for partial-fraction expansions, and we discuss
procedures for (a) simple roots, (b) complex conjugate roots, and

(c)

multiple roots.

should be recalled that if the degree of the numerator is greater or

equal to the degree of the denominator, we can divide the numerator by
the denominator such that the remainder can be expanded more easily
It

into partial fractions.

Consider the following example:

m=

D(s)
Since the degree of N(s)

is

s'

3s*

+ 3s + 2

2s

greater than the degree of D(s),

we

divide D(s)

into N(s) to give

F(s)

'

= 5 + 1s

+2s +

(6.98)

Here we see the remainder term can be easily expanded into partial
However, there is no real need at this point because the
denominator s2 +^2s + 2 can be written as

fractions.

s*

+ 2s +

= (5 +

1)*

(6.99)

We can then write F(s) as

so that the inverse transform can be obtained directly from the transform
tables,

namely,

C-W)] =
From this
form

example,

<5'(0

3(0

+ er*(sm t - cos i)

we see that intuition and a knowledge of the trans-

table can often save considerable work.

examples in which intuition plays a dominant

Example

6.11.

(6.101)

Find the

Consider some further

role.

partial-fraction expansion of

2*

+3

(JTWTT)

(6102)

The
If we see that F(s)

(s

Find the

+ 1) + (s + 2)
+ Dfr+g

(6103)

fr

then the partial-fraction expansion

6-12.

149

can also be written as

*W-

Exanpfe

Laplace transform

is trivially

partial-fraction expansion

of

s +5
j
+5
F(s) = , T -,

(6-105)

We see that F(s) can be rewritten as

= _i_ T _i_^
mW = feai
+
+
+
2)

(s

(s

(6 106)

2)

(s

2)

Real roots

Now let us discuss some formal methods for partial-fraction expansions.

First

we examine a method

for simple real roots.

m=

- s tt^t:
)(s s0(s St)

(6-107)

(s

where s< st, and st are distinct,

Expanding F{s) we have
F(s)

Consider the function

real roots,

and the degree of N(a)

= -*- + -^- + -^S Si

S Si
S S

Let us first obtain the constant A,.

j ) to give
of the equation by (s

< 3.

(6.108)

We proceed by multiplying both sides

(s

s)F(s) = K

H
s

If we let 5

=*

in Eq. 6.109,

we

obtain

Ji-(#-J)J^)|^
Similarly,

we

(6.109)

s,

see that the other constants can

(6110)

be evaluated through the

general relation

A,
1

6.13.

= (*-*) ft*) |_.,

(6.111)

Let us find the partial-fraction expansion for

J +2j - 2
.g + J^- + J^_
FV- j(*+2X*-3)
FW
*
s+2+3-3
'

Z)
(6.112)
K

150

Network

anal/sis

and synthesis

Usin gEq. 6.111, we find

Ao

sF(s) |_>

+ 2s -2
+ 2X* - 3)

s*

(*

-o

+2s -2
-3) a

s*

Kl

+ 2s - 2

s2

*"

(6.113)

sis

5
13

~15
3

s(s+2)

Complex

roots
Equation 6.111 is also applicable to a function with complex roots in
its denominator. Suppose F(s) is given by

F(s)

JVC*)

D&Xs -

Ki

s-x-jfi
where A^/Di

is

- JPK' - * + JP)
+

the remainder term.

K1 =

JVi(s)

s-x+jp

(6.114)

Djis)

Using Eq. 6.111, we have

ypDM+m
(6.115)

-irfD^-m
where we assume that j = a jP are not zeros of Z>i(s).
It can be shown that the constants Kx and K% associated with conjugate
roots are themselves conjugate. Therefore, if we denote K as
(6.116)

K2 = A-jB = K *

then
If

we denote the
we see that

inverse transform of the

complex conjugate terms as

fx(t),

A(0
i(o

(6.117)

_i

e"'(K ie "

2c**(A cos Bt

Kft-"*)

- B sin pi)

(6.118)

The Laplace transform

151

A more convenient way to express the inverse transform f^t) is to introduce the variables

M and

denned by the equations

<X>

= 2A
McosO = -25
where

terms of

and

M and

obtain

the inverse transform

<I>,

When

we note

lt

6.116.

In

(6. 120)

that

M cos O + jM sin O =

Mf =
6.14.

Kr in Eq.

is

-2B + ;2,4 = 2;XX

related to the original function F(s),

Example

(6.119)

= Me?* sin (0t + O)

M and $ from K

Me'*

<1>

and imaginary parts of

are the real

/i(0

To

sin

we

see

(6.121)

from Eq. 6.115 that

*<+Jfl

(6 . 122)

Let us find the inverse transform of

w-vr^rkr+ii
For the simple root s

2, the constant

(6123)
.

K is

K = (*+2)F(*)|,__8 =

(6.124)

For the complex conjugate roots

**+2j+5=(* +
We see that

= 1,
Me**

The inverse transform

C-i[F(*)]

+j2Xs

-y'2)

(6.125)

= 2, thus
*

+3

2(*+2) i1+*2
is

V5

c-i(tan-j+W)

(6.126)

then

- e-**

+ = <r* sin \^

~\~ tan-1 2j
(6.127)

= - e~u
5

2
-p e~*

V5

Multiple roots
Next let us consider the case

-1
tan 2)

which the partial fraction involves

examine two methods. The first
the second does not.

repeated or multiple roots.

requires differentiation;

cos (2/

We

in

will

Network

152

analysis

and synthesis

Method

Suppose we are given the function

W=

F(S)

(6.128)

- s )" ^(s)
The partial fraction expansion
with multiple roots of degree n at s = j
(s

of F(s)

is

Kt

Kx

(s-s ) n

(s-so)""

..... *-i

s-s

(s-s )-8

*i(?)
J>i()

(6.129)

where N^jDjUs) represents the remaining terms of the expansion. The

problem is to obtain K^K^..., K^. For the K^ term, we use the method
cited earlier for simple roots, that

Ko
However,

if

we were

is,

= (s-s rF(s)\

(6.130)

to use the

same formula

to obtain the factors

K ,..., #_!, we would invariably

t

dition.

Ku

arrive at the indeterminate 0/0 conn

s ) and
Instead, let us multiply both sides of Eq. 6.129 by (s

define

F1(s)s(s-s rF(s)

(6.131)

Thus

Fx(s) -

A,

+ Kt(s - s ) +

+ K^s - So)""1 + *(*)(' - s*Y

(6.132)

where R(s) indicates the remaining terms. If we

s,

we

differentiate

Eq. 6.132 by

obtain

Fi(s) - *i + 2K^s 4
as

s )

*_!(/!

D(s

- so)""* +
(6.133)

Kx = Ft(s)

It is evident that

(6.134)

ds

On the same basis

and

Kt =

F^s)

(6.135)

2 ds

in general

K = -tWs)!
!->
.

j\ ds'

= 0,1, 2

-1

(6.136)

The
Example

6.15.

Laplace transform

153

Consider the function

(6 - ,37)

'W-STTlJi
which we represent-in expanded form as

-(7Ti? + (JTI? + 7TT + 7

The constant A

for the simple root at s

<'>

is

A=sF(s)\^=-2
To obtain

the constants for the multiple roots

Fjis)

= (s +

l)

(6.139)

we first find Ft(s).

8 F(s)

-^

(6.140)

Using the general formula for the multiple root expansion, we obtain

^i=T7T
1 ds
!

fs

-2\

!_!

,_,.

=2
-3
|__!
I

<6142)

*-i^)L-(-5)L-;
IW __2 gl +

sothat

i+

-i

-5

(6.144)

Method B
for arriving at the partial-fraction expansion for

The second method

no differentiation. It involves a modified power

Let us consider F(s) and F^s), defined in Eqs. 6.128
s
Joand 6.131 in Method A. We define a new variable/; such that/>

multiple roots requires

series expansion.*

Then we can write F^s) as

*(p
Dividing N(p

+s)

ascending powers ofp,

fiG*

+ tf-?T^

by

D%(p + s

we

obtain

),

(6145)

with both polynomials written in

+ s ) = K + K p + Ktpt +-- +
1

X^j,""1

f"f
+ *o)

0l(l>

(6.146)
* I

am

Institute

indebted to the late Professor Leonard O. Goldstone of the Polytechnic

of Brooklyn for showing

me this method.

Network

154

The

anal/sis

and synthesis

original function F(s)

F(P

o)-

is

F (p + s ) by the equation

related to

-,-%-!+

pn

Dl(p

+ So)
(6.147)

Substituting 5

F(s)

= p in Eq. 6.147, we obtain

* +

g
/

\ Bl +

-^ + -\

(6.148)

We

have thus found the partial-fraction expansion for the multiple-root

terms. The remaining terms KjlD^s)] still must be expanded into partial
fractions. Consider Example 6.16.
Example 6.16

*wUsing the method just given, F^s)

F^s)
Setting p

=s+

1,

we

(s

fr

(6149)

ijUa

is

+ If F(s) =

-^

(6.150)

then have

- 1) into a series as given in Eq. 6.146 requires dividing

the numerator 2 by the denominator 1 + p, with both numerator and denominator arranged in ascending power of p. The division here is
The expansion of Ft(p

- lp + 2/

l+/0"2
2 +2/>

-2^

-2p-2f

Since the multiplicity of the root

we have three terms in the quotient.

AT = 3, we terminate
We then have

is

the division after

FX{P -l)=2-2p+2p*- j-x

The original function F(p 1) is
F(P

1)

(d -1)
=-FJx7T- -

? ?

(6.152)

+~ -pTl
p

(6153)

The Laplace transform

Substituting s

= />, we have

m
Example

- (7TT? " (7TI? + (7TT) " JT2

As a second example,

6.17.

F(,)

Since

we have two

sets

= 0;

(6155)

=^T1?
= 0, and s =

1) we have a
Let us arbitrarily choose to
s here, we do not have to make any substitutions.

since /j

(6154)

consider the function

of multiple roots here

F&) is then

(at s

Ooi s = 1.

choice of expanding F^s) about s

expand about *

155

2+s

=(7TT? = rT2FT7
Ft(j), we obtain

Expanding

F^s)

is

then

F(s)

+ 4)
^ + iy
3* + 4
+
+^

- 3j +

2
2

F(s)

-^ -

We must now repeat this process for the

Fortunately

we

3s

answer

POLES

(6.158)

term

1)*

is

+4 = 3Qr +

1)

(*

1)

iy

1
3_ +
+ I (s +

(6159)
1)

then

"W-l-i+TTT + frTi?
6.6

(6.157)

see that the term can be written as

(s
final

s\3s

+4

3*
(s

The

<6156>

(6160)

AND ZEROS

In this section

we

will discuss the

many

implications of a pole-zero

description of a given rational function with real coefficients F(s).

We

be the roots of the denominator of F(s). The

zeros of F(s) are defined as the roots of the numerator. In the complex s
plane, a pole is denoted by a small cross, and a zero by a small circle. Thus,
define the poles of F(s) to

for the function

_ ^-i+;D(s-i-Jl)
+ l)\s+j2)(s-j2)

s =
1
(double)
s = y*2
(s

the poles are at

+y*2

156

Network

and synthesis

analysis

jo
|

plane

j2 i

.t

-<r

,1

i_W_

Double

pole.*.

12

-3 -2 -1

->2 x
-J<0

FIG.

and the zeros are

diagram of F(s).

6.7. Pole-zero

at

00

-j\

zeros of F(s) are shown in Fig. 6.7.

Now let us consider some pole-zero diagrams corresponding to standard
example, the unit step function is given in the complex

The poles and

signals.

For

frequency domain as
C[u(0]

=-

6162>

and has a pole at the origin, as shown in

signal e*, where <x > 0, has a transform

function cos

single pole at s
<a

i,

C[coso,

(6.163)

Oq

-o-* as indicated in

whose transform

The exponential

C[e']
which has a

Fig. 6.8a.

Fig. 6.86.

The

cosine

is

t]

= -r-1-~t

(6-164)

7<u
poles at s
has a zero at the origin and a pair of conjugate
diagram
pole-zero
the
shows
6.Sd
Figure
as depicted in Fig. 6.8c.
is
corresponding to a damped cosine wave, whose transform
Z[e-"* cos

co t ]

g
= s +r=-*
(s +
r+
<r

(6.165)
o>

The

Laplace transform

157

I
(a)

(b)

]W

m)'
""

M,

+,

*******

-a
x

-jao
id)

FIG.

6A

Poles and zeros of various functions.

From these four pole-zero diagrams, we note that the poles corresponding to decaying exponential waves are on the a axis and have zero
imaginary parts. The poles and zeros corresponding to undamped
on the/o> axis, and have zero real parts. Consequently, the
and zeros for damped sinusoids must have real and imaginary

sinusoids are

poles

parts that are both nonzero.

Now let us consider two exponential waves/x(0 = er'i* and/i(f) = er***,

where at
6.9a.

>

<r

> 0, so that/,0) decays faster than/iO), as shown in Fig.

The transforms of the two

functions are

*i()

=
s

ax

(6.166)

F&) =
s

a,

by the pole-zero diagram in Fig. 6.96. Note that the further

the pole is from the origin on the a axis, the more rapid the exponential
decay. Now consider two sine waves, sin atjt and sin co t t, where <o t >
>! > 0. Their corresponding poles are shown in Fig. 6.10. We note here
as depicted

JU>

s plane

*2(K

Fi(K

0"1

V2
(a)

FIG.

*.*. Effect

(b)

of pole location upon exponential decay.

Network

158

analysis

and synthesis

JU
1

X./C02
JO)

X
,

>i

X
1

a
X
si*
(

-jwi

X
2*

<

-j(i>2

FIG.

FIG. 6.10. Pole locations corresponding to sin to,/ and sin mj.

6.1

represents frequency of
that the distance from the origin on theyco axis
frequency.
oscUlation; the greater the distance, the higher the
responses fx {f) and
time
the
compare
let
us
thumb,
of
rules
these
Using
shown in Fig.
corresponding to the pole pairs {* lf s 1 *} and fo, s a *}
a (0
sinusoids.
damped
see that both pairs of poles corresponds to

6.11.

We

than/a(f)
sinusoid/i(0 has a smaller frequency of oscillation
s t Also,
of
part
because the imaginary part of * x is less than the imaginary
time
The
Res
because
2
1
(t) decays more rapidly than fx (t)

The damped

fz

responses /x(0

>Kcs

and/a(0

are

shown

in Fig. 6.12.

f2 (t)

- lme*

h) =
\^" ~~-

C-

lme'

___ __
-

=*-

S ^-^
(b)

in Fig. 6.11.
FIG. 6.12. Time responses for poles

The Laplace transform

159

ju
plane

m
at

-m

(b)

FIG. 6.13. Effect of right-half plane poles upon time response.

Let us examine more closely the effect of the positions of the poles in
upon transient response. We denote a
pole pt as a complex number (
at +jco i For a given function F(s)
p
with only first-order poles, consider the partial-fraction expansion
the complex frequency plane

F(s)

The

_*!L_

s-

+ _*!_ +

(6.167)

s-P

-Pi

Po

inverse transform of F(s)

/(*)

is

= K e + Kie* +
= K^"*e im,t + K1e'

+K
lt

iait

ne"*

+ + K^e*""*

In/0), we see that if the real part of a pole is positive, that

the corresponding term in the partial-fraction expansion

K e"
i

*e

is,

at

(6.168)

> 0, then

iatt

an exponentially increasing sinusoid, as shown in Fig. 6.13a.

thus
see that poles in the right half of the s plane (Fig.
6.13*) give rise to
exponentially increasing transient responses.
system function that has
poles in the right-half plane is, therefore, unstable. Another
is

We

unstable

situation arises if there

is

a pair of double poles on theyw

axis,

such as for

the function

F(

"-(7f^'

whose pole-zero diagram

F(s)

is

shown

in Fig. 6.14a.

(6.169)

The inverse transform of

is

/(0

= - sin co
2io

(6.170)

160

Network

analysis

and synthesis

Vjao

*-.7o

f(t)

m ^-

sin (dot

(b)

(a)

FIG. 6.14. Effect of double zeros on

they'd) axis

upon time

response.

which is shown in Fig. 6.146. It is apparent that a stable system function

cannot also have multiple poles on theytu axis.

N(s)lD(s).
Consider the system function H(s)
numerator and denominator polynomials, we obtain

H(s)

- Zq)(s - Zi)
- P )(S - Pi)

Hg(s
(S

If

- g)
(S
Pm)

(s

we

factor the

(6.171)

H(s) is completely specified in terms of its poles and zeros

and an additional constant multiplier #. From the pole-zero plot of H(s),
obtain a substantial amount of information concerning the

It is clear that

we can

whether the
behavior of the system. As we have seen, we can determine
the jm axis
and
poles
for
plane
right-half
the
checking
by
system is stable
its transient
concerning
information
obtain
We
poles.
for multiple
discuss
behavior from the positions of its poles and zeros; and, as we will

diagram also gives us significant.information

We thus
concerning its steady-state (Jm) amplitude and phase response.
see the importance of a pole-zero description.
Suppose we are given the poles and zeros of an excitation E(s) and the
response
system function H(s). It is clear that the pole-zero diagram of the
and
H(s)
of
diagrams
pole-zero
the
of
R(s) is the superposition
in Section 8.1, the pole-zero

function
(j).

Consider the system function

W"
and the

(6.172)
(i

- ftK " A)

excitation
s

(6.173)

pt

The Laplace transform

Then

the response function

H(s)

161

is

- Zq)(s - i)
- p )(s - pj(s - p)

gpEofr
(s

(6.174)

R(s) contains the poles and zeros of both H(s) and (*),
except in the case where a pole-zero cancellation occurs. As an example,
let us take the system function
It is clear that

H(s)

2(5
(s

2+j4)(s

1)

(6.175)

+ 2-j4)

whose pole-zero diagram is shown in Fig. 6.15c. It is readily seen from the
pole-zero diagrams of the excitation signals in Fig. 6.18a that the step
response of the system has the pole-zero diagram indicated in Fig. 6.156.

The response to an
tion of Fig.

we

zero plot,
plot

6. 1 5c.

excitation signal 3 cos It has the pole-zero representa-

To specify completely the given function F(s) on a pole-

indicate the constant multiplier of the

numerator on the

itself.

Let us examine the significance of a pole or zero at the origin. We

that dividing a given function H(s) by s corresponds to integrating
the inverse transform h(t)
C- l [//(*)]. Since the division by s corresponds

know

to a pole at the origin,

we

see that a pole at the origin implies

an

inte-

gration in the time domain.

Because the inverse transform of H(s) in

Fig. 6.15a is the impulse response, placing a pole at the origin must
correspond to the step response of the system. In similar manner, we
deduce that a zero at the origin corresponds to a differentiation in the
time domain. Suppose we consider the pole-zero diagram in Fig. 6.15c
with the zero at the origin removed; then the resulting pole-zero plot is
the response of the system to an excitation E sin It. Placing the zero at
the origin must then give the response to an excitation

Et cos It, which, of

ju

J(0

ju

J*

K-2

JT-2

JT-6
j2::

^r

-2 -1

-2 -1
-j2x

-;4-

(a)

FIG. 6.15. (a) System function,

excitation e(j)

= 3 cos 2i.

-74

-74-

(b)
(jb)

Response to unit step

(0
excitation,

(c)

Response to

162

Network

course,

is

true.

differentiator

analysis

We

and synthesis

therefore conclude that the system function of a

must have a zero

have a pole at the

an integrator must

at the origin; that of

origin.

EVALUATION OF RESIDUES

6.7

Poles and zeros give a powerful graphical description of the behavior

of a system. We have seen that the poles are the complex frequencies of
the associated time responses.

What

role

this question, consider the partial fraction

where we
s

shall

do the zeros play? To answer

expansion

Si

assume that F(s) has only simple poles and no poles

at

oo.

The

inverse transform

is

/(0

fV

(6-177)

time response /(r) not only depends on the complex

on the constant multipliers t These constants t
are called residues when they are associated with first-order poles. We will
show that the zeros as well as the poles play an important part in the

It is clear that the

frequencies s( but also

determination of the residues

Earlier

we

discussed a

number of

methods for obtaining

different

the residues by partial-fraction expansion. Now we consider a graphical

method whereby the residues are obtained directly from a pole-zero

diagram. Suppose

we

f(s)

are given

= 4>(8 ~ goX5 ~ ^
(s - p )(s - Pi)

where

m>

n and

all

F(s)

Our

task

is

(s

( 6 .i78)

the poles are simple. Let us expand

-^ +

= -&- +
S Po
*

Pi

to determine the residues A,.

^ = (s-ft)^(s)

~ 2 )
- pj

(s

F(s)

as

+ -=* Pm

(6-179)

We know that

Zq)(p< - zi) (Pi - z)

(Pi - Pm)
(Pi ~ Po)-" (Pi - Pi-^iPi - ft+l)
A

(Pi

'

'

(6.180)

The Laplace transform

When we interpret the Eq. 6.180 from a

complex-plane viewpoint, we see that each one
z f) represents a vector drawn
of the terms (p (

JU
ozo

Pix

from a zero

z t to the pole in question,

Similarly, the terms (p f

where

/>),

163

t.

^ k,

from the other poles to the

pole p^ In other words, the residue t of any
represent vectors

P0

pole

is

equal to the ratio of the product of

the vectors from the zeros to

t,

to the prod-

uct of the vectors from the other poles to

To

illustrate this idea, let

{.

us consider the pole-

A*

*1

FIG. 6.16. Poles and zeros

of F(s).

zero plot of
F(s)

=
(S

given in Fig. 6.16.

The
F(s)

Ms - zp)(s - Zl)
- p )(s - p )(s - ft*)

partial-fraction expansion of F(s)

- K
s

(6.181)

Kl
Pt,

pt

is

^1

(6.182)

-Pi-

where the asterisk denotes complex conjugate. Let us find the residues A"
and Kx by means of the graphical method described. First we evaluate x
by drawing vectors from the poles and zeros to/^, as shown in Fig. 6.17a.

The

residue

x is

then

K1 _

XqAB
(6.183)

CD

where symbols in boldface represent vectors. We know that the residue

of the conjugate pole^* is simply the conjugate of x in Eq. 6.183. Next,

FIG. 6.17. Determining residues by vector method.

Network

164

and synthesis

analysis

to evaluate

ju

C-3

K& we draw vectors from

poles and zeros to

/>

We see that

6.176.

K - A RL

(6.184)

MN

-,

the

as indicated in Fig.

With the use of a ruler and a protractor,

determine the lengths and the angles of

-1

we

the vectors so that the residues can be determined quickly and easily. Consider the fol-

-jl

lowing example. The pole-zero plot of

55

F(s)
FIG. .I8. Pole-zero diagram

of-

2)(s

+ 1 - jl)( + 1 + jl)

F(s).

is

shown

(6.185)

The

in Fig. 6.18.

F(s)

at

1 +jl

are

is

Kt*

=
s

First let us evaluate

partial-fraction expansion of F(s)

"*"

+ jl

(6.186)
s

+2

The phasors from the poles and zeros to

x.

shown

+ - jl
1

in Fig. 6.19a.

Kx =

X -p

We see that Kt is
^2/135

J2 1*5_ X

2/90

_3
2

JU

C-3

AIVr

-2/\&
-<r

180*

nJ

'

V90*
r

(a)

-a

(b)

FIG. 6.19. Evaluation of residues of F(s).

the pole

The

From

Fig. 6.19*

we find the value of the

3

^_
K,=!

Laplace transform

x 2_

= _3

V2 /-13S x 72 7+135

+ l-jl

THE

6.8

In

INITIAL

this section

transforms.

The

value of /(/) at
infinity,

that

AND

we

is

+ l+jl

(6.187)

+2

FINAL VALUE THEOREMS

will discuss

two very

useful theorems of Laplace

value theorem.

It relates the initial

first is

the

= 0+

to the limiting value of sF(s) as s approaches

initial

is,

lim f(t)

The only

Kt to be

residue

Therefore, the partial-fraction expansion of F(s)

F(s)

165

restriction is that /(f)

= lim s F(s)

(6.188)

must be continuous or contain,

at most,

step discontinuity at / == 0.

In terms of the transform, P(s)

t[f(t)];
this restriction implies that F(s) must be a. proper fraction, i.e., the degree
of the denominator polynomial of F(s) must be greater than the degree of
the numerator ofF(s). Now consider the proof of the initial value theorem,

which we give in two

(a)

From

parts.

The

function /(f)
the relationship

Cf/'(0I

we

obtain

Therefore

continuous at

is

= 0,

that is,/(0 )

=/(0+).

=/ V(0r" dt = s F(s) - /(0-)

hm C[/'(01 -

lim s F(s)

-oo

- /(0-) =

(6.189)

(6.190)

Um s F(s) = /(0-) = /(0+)

(6.191)

-oo

(b)

The function/(0 has a step discontinuity

at

= 0.

Let us represent

f{t) in terms of a continuous part/x(0 and a step discontinuity

as shown in Fig. 6.20.
can then write /(/) as

D (/),

We

/(0-/i(0
where

D =/(0+) /(0-).

The

+ J>W

derivative of/(f)

f'{t)=f\(f)

Do{t)

(6^2)
is

(6.193)

Network

66

anal/sis

and synthesis
Du(t)
fi(t)

fit)

FIG. 6.20. Decomposition of a discontinuous function into a continuous function

plus a step function.

Since fx (t)

is

continuous at

we know from

0,

lim s F^s)

= A(0-) = /(0-)

Taking the Laplace transform of both

s F(s)

which

simplifies to

part (a) that

of Eq. 6.193,

sides

(6.194)

we have

- /(0-) = s F^s) - M0-) + D

s F(s) <= s F (s) + D
t

(6.195)
(6.196)

Now, if we take the limit ofEq. 6.196 as s-+ oo and let/(0+) -/(0-) =
D we have
(6.197)
lim s F(s) = lim s F x (s) + /(0+) - /(0-)
By Eq.

6.194,

we then

obtain

limsF(s)=/(0+)
Example

6.18.

(6.198)

Given the function

m=

2(s
s*

1)

+2s +

(6.199)

us find/(0 -f ). Since the degree of the denominator is greater than the degree
of the numerator of F(s), the initial value theorem applies. Thus

let

lim s F(s)
Since

2(s

\)s

lim

C-H^Cj)]

=2

=2c-*cos2r

(6.200)
(6.201)

we see that /(0+) =2.

Example

6.19.

Now let us consider a case where the initial value

theorem does

not apply. Given the function

f(t)

we see that/(0+)

3.

(5(0

+ 3r*

The transform of/(f)

F(s)

(6.202)

is

+
s

(6.203)
1

The Laplace transform

lim sF(s)

so that

= lim si 1 +
.-co

*-*,

oo

Next we consider the final value theorem, which

(6.204)

/
states that

= lim s F(s)

lim /(0

167

(6.205)

provided the poles of the denominator of F(s) have negative or zero real
parts, i.e., the poles of F(s) must not be in the right half of the complexfrequency plane. The proof is quite simple. First

f'(t)e-' dt

- /(0-)

s F(s)

i:
Taking the

limit as s

->

in the Eq. 6.206,

/'() dt

lim

j;
Evaluating the integral,
/(oo)

(6.206)

we have

s F(s)

- /(0-)

(6.207)

-o

we

obtain

- /(0-) - hm 5 F(s) - /(0-)

Consequently,

/(oo)

= lim s F(s)

(6 .208)

(6.209)

-o

Example

6.20.

Given the function

/(/)

3(/)

+ 2e~*

(6.210)

which has the transform

F(s)

let

= iii
*(s + l)

=l +

us find the final value/( oo). Since the poles of F(s) are at s
the final value theorem applies. We find that

(6 . 211)

=0 and s = 1

we see

]imsF(s)=3
which

is

the final value of /(/) as seen

Example

6.21.

we see that

Given

from Eq.

f(t)

lim/(f)

But from

sF(s)

= le*
= oo

lim * F(s)

6.210.
(6.213)
(6.214)

7s

=
s

we have

(6.212)

(6.215)

(6.216)

-o

We see that the final value theorem does not apply in this case, because the pole
= 1 is in the right half of the complex-frequency plane.

Network

168

analysis and synthesis

TABLE

6.1

Laplace Transforms
F(s)

fit)

1.

FW-JT/Oy**

fit)

2.

i/i(0

3.

|/W

4-

^/<

+ ./)

CxFxC^+fljF^)

fW

-/(0-)

dn

5.

a" Fis)

f/W

!*w

rfT

6.

f
f fir)drda
Jo- Jo-

7.

(-0"/W

8.
9.

10.

fit -a)u(t

- a)

>(,)

e-F(*)

**/X0

F(s

8(t)

- a)

dn
1

12.

(/)

13.

14.

-r

s
1

n\
1

15.

16.

/?- (

17.

sin cof

*
1

e-

- J f^'f^O-)

a
1

_ e-ft)
a>

The Laplace transform

TABLE

6.1 (cont.)

s
18.

cos

19.

sinh at

20.

cosh a/

21.

"'sin or

22.

*""' cos

cor

+ <0
a

i*-fl
0>

(*

(5

(* + )
+ )* +

mt

CO

co

e-**t*
23.

(j

24.

25.

/(/); n -0,1,2,3,...

*)

a)**"1

*
.

sin co/

2co

0*

<*)

(5*

a)W[(i

(Bessel function of first kind,

nth order)
26.

J-

(t/)-*

T(*
27.

r*

(A:

1)

need not be an integer)

Problems
6.1

Find the Laplace transforms of

(a)

/(0-sin(orf+|8)

(6)

/(/)

(*)

= *-<*> cos (at + P)

f(t) .- (r + 1>-"
A" is a real constant
/(/) - JT*;
/ - (** + - 0^**

(/)

f)-td'(t)

(g)

/W - aW* - 4)]

(c)
(</)

6 J,

>1.

Find the Laplace transforms of

- cos (t -

*/4) u(f

(a)

/(/)

(6)

/(0-cos(/-W4)(0

w/4)

a)H

- 5]-

169

170
6.3

Network

analysis

and synthesis

Find the Laplace transforms for the waveforms shown.

Kt)

fit)

-1
(a)

6.4

(b)

Find the Laplace transforms for the derivatives of the waveforms in

Prob. 6.3.
6.5

Find the inverse transforms for

(a)

F(s)

(6)

{S)

2s

=
(s

"

+9

3X*

+ 4)

5s - 12
+ 4s + 13
4* + 13

s*

GO
all the inverse transforms may be obtained without resorting to
normal procedures for partial-fraction expansions.

Note that

The

Laplace transform

171

Find the Laplace transform of the waveforms shown. Use (a) the infinite
(6) the Laplace transform formula for periodic waveforms.
Both answers should be in closed form and agree.
6.6

series

method and

fit)
fit)

uv\\A

3T

2T

2T

3T

-A
lb)

la)

*t)

+1

r/
2

>

/
2T

-1

PROB. 6.6

(e)

6.7

Id)

Evaluate the definite integrals

w
t

()

sin It dt

6.8

Solve the following differential equations using Laplace transforms

(a)

x'{t)

(b)

x'(t)

(c)

x"{t)

id)

x"{f)

+ fceXO + 9x(t) = cos 2t

+ 3x-(/) + 2a<0 = (5(0
+ 2^(0 + 5a<0 = u(t)
+ Sx'(t) + MO = {e~* + <r**) (0

(e)
It is

sj-CO

given that

6.9

cos 3f dt

a<0 )

= as'fl) )

+ 2as'(0 = (0 + *~* (0
= for all the equations.

Using Laplace transforms, solve the following

sets

equations:
()

2*'(0
x'(0

(b)

2x'{t)

+ 4a<0 + y'(f) + 2y(t) = <J(0

+ 3a</) + y'(f) + 2y(r) =
+ (/) + y'{t) + 4jK/) = 2e-

+ jr*0) +
The initial conditions are all zero at t = 0
*'(0

8y(0

(5(0

of simultaneous

172
6.10

Network

anal/sis

and synthesis

Find the inverse transforms for

H0-

()

+ 9X' +
+a

(s*

j
()

3)

s+P
(*+!)

(c)

S*

s*+2s +2

tf>

+ 3* + 1
s* +s
s + 5
~(s + W* + 2)
*

to

F(s)-

(/)

ns)

(g)

F(s)

(A)

F(s)~

s*

6.11

+ lX*+2)*
3j - J* - 3s + 2
(*

sKs-W

Find the inverse transforms for

+-*'

()

sH.s+2)

_ r-s
+ 3s + 2

->*

(ft)

F(')

(c)

*"(*)

se
**

re-"
s

+0

write
6.12 Given the pole-zero diagrams shown in (a) and (b) of the figure,
What
can
polynomials.
of
quotients
Fjs)
as
and
(s)
the rational functions t
you say about the relationship between the poles in the right-half plane and the
signs of the coefficients of the denominator polynomials?

FiM

ju

-A

A
>ij0.6

Fzb)
*

>>0.6

-1

+1
H-/D.6

|->D.6

-A

-A

to

PROS. 6.12

The Laplace transform

173

6.13 For the pole-zero plots in Prob. 6.12, find the residues of the poles by
the vector method.

6.14

For the pole-zero

plots

shown

in the figure, find the residues of the

poles.

-2

j2

jl<

<

-1
-jln
-J2

(a)

(b)

ja

f-H n
-<r-3

-2

-x-

-1

-1

+1

PROB. . 14

6.15 Two response transforms, R^s) and R/s), have the pole-zero plots
shown in (a) and (b) of the figures, respectively. In addition, it is known that
i(0 +) - 4 and /-,(/) - 4 for very large t. Find r^t) and rjf).

Sd)

ja

X
I
1

-V -2

-l

6
-1?

i:

'

(a)

-n

(b)

PROB. 4.15

Network

174
6.16

It is

analysis

and synthesis

given that
_.
F(s)

as*

+ bs + c

+ 2j* +s +

s*

and c such that/(0 +) =/'(0 +) = /'(0 +) = 1

Using the initial and final value theorems where they apply, find/(0 +)
and/(oo) for

Find

a, b,

6.17

(6)

F(*)

3X*

(*

IX*

(c)

1)

s* + 3s + 2
FW - * + 3* + 3j +
2

5<s

<

+ 4)

+ 4X* +

8)

^-"cr+ixr+Q

chapter

Transform methods

network

7.1

in

analysis

THE TRANSFORMED CIRCUIT

In Chapter 5 we discussed the voltage-current relationships of network

elements in the time domain. These basic relationships may also be
represented in the complex-frequency domain. Ideal energy sources, for
example, which were given in time domain as v(t) and i(f)> may now be
represented by their transforms V(s)
resistor, defined

by the

(0
is

= [t<01

and

I(s)

= Cfl(f)].

*(')

denned in the frequency domain by the transform of Eq.

V(s)

For an inductor, the

The

v-i relationship

RI(s)

7.1,

71 )

or
(7.2)

defining v-i relationships are

w- !
1

(7.3)

(0

= 7p

Transforming both equations,

F(s)

/(5)

we

t<T)dT-M(0-)
obtain

= sLI(s) - Li(0-)
=

J-F(S)
sL

+ ^;

i(O-)
s

175

CM)

176

Network

analysis and synthesis

IM
IM

f\HO-)

VM

)i(0-)

FIG.

7.1. Inductor.

The transformed
Fig. 7.1.

circuit representation for an inductor

For a capacitor, the defining equations are

v(t)

=-

'

i(r)

C Jo-

dr

is

depicted in

+ v(0- )
(7.5)

*>- c

The frequency domain counterparts of these equations

are then

F -J; / + *&=>
sC

I(s)

sC

V(s)

(7. 6)

- C o(0-)

as depicted in Fig. 7.2.

From this

analysis

we

see that in the complex-frequency representation

and admittances in
For
example,
series or parallel with energy sources.
from Eq. 7.4, we see
that the complex-frequency impedance representation of an inductor is sL,
and its associated admittance is XjsL. Similarly, the impedance of a
capacitor is 1/sC and its admittance is sC. This fact is very useful in circuit
analysis. Working from a transformed circuit diagram, we can write
mesh and node equations on an impedance or admittance basis directly.

the network elements can be represented as impedances

-%

IM

IM

VM
+

=PC (l)

K0-)

FIG. 7.2. Capacitor.

V<

Transform methods

in

network

analysis

177

vs(t)

vi(t)

FIG. 7.3

The process of solving network differential equations with the use of

transform methods has been given in Chapter 6. To analyze the circuit
on a transform basis, the only additional step required is to represent all
the network elements in terms of complex impedances or admittances with
associated initial energy sources.
Consider the example of the transformer in Fig. 7.3. If we write the
defining equations of the transformer directly in the time domain, we have

(7.7)

Transforming

this set

of equations, we obtain

V&) = sLy I^s) V&) = sMIM -

+ sM Is) - M ijp-)
M ^(0-) + sL, Us) - U i,(0-)

Ly i^O-)

(7.8)

This set of transform equations could also have been obtained by representing the circuit in Fig. 7.3 by its transformed equivalent given in Fig. 7.4.
In general, the use of transformed equivalent circuits is considered an
easier

way

to solve the problem.

v2W

FIG. 7.4

178

Network

analysis

and synthesis

-nnnp
fi,(0-)

+__
+

12

BC (0-)

FIG. 7.5

Example 7.1. In Fig. 7.5, the switch is thrown from position 1 to 2 at t = 0.

Assuming there is no coupling between L^ and Z*, let us write the mesh equations
from the transformed equivalent circuit in Fig. 7.6. The mesh equations are

-.+.)

*->
-l
+ A'i,(0-) = -^/

1(5)

'm^Qi
^

I2

Li

(7.9)

^>

vuo'

L1M0-)

1*1140-)

QviW

FIG. 7.6

Example
f

= 0.

7.2.

In Fig. 7.7, the switch

Just before the switch

amp, o(0 )

= 2 v.

is

is

thrown from position

thrown, the

Let us find the current

initial

i(t)

Qv(t)

>'

FIG. 7.7

to 2 at time

after the switching action.

lh

-x

conditions are /^(O )

=2

Transform methods
3

ivw

network

179

anal/sis

=0

nnnp-

in

/i\

FIG. 7.8

Since the switch

is

closed at

formed

= 0, we can regard the S-v battery as an equivacircuit is

The mesh equation

circuit.

Solving for

The

lent transformed source 5/s.

I(s),

+2

now redrawn

in Fig. 7.8 as

for the circuit in Fig. 7.8

(3

trans-

is

+s+ ?) 70)

C7.10)

we have

+3
1
(7.11)
+ IX* + 2) s + 1 s + 2
i(0 = C-HAO] = e~* + -
(7.12)
Consider the network in Fig. 7.9. At t = 0, the switch is opened.
2s

(j

Therefore,

Examine 73.

Let us find the node voltages

vjCf)

and

v%(t) for the circuit. It

Z,=Jh
G = 1 mho

C=
V=

is

given that

lf
1

L
i<0-)

+
C=fcoc (0-)

FIG. 7.9

Before

we

substitute element values, let us write the

node equations for the

transformed circuit in Fig. 7.10. These are

NodeV1

*_

_ (l(0-0 +
Node

Vt

CPc(o-)

/
\
= (sc
+ -)ki(o
1

-- vM

^-4 ()

(7.13)

Network

180

anal/sis and synthesis

V2

FIG. 7.10
If

values into the set

(t(0 ) =
of node equations, we have
1

had been in steady

amp. Substituting numerical

the switch opening the circuit

we assume that prior to

then we have v^O-)

state,

1 v,

(*+^K (i)-?K/f)

(7.14)

Simplifying these equations,

s

we obtain

1
1

= (s* + 2)Vt(s) - 2V&)

. -2Vx{s) + (s + 2)Vs)

Solving these equations simultaneously,

we have

J+l
s +1
+ 2s + 2 " (s + l)1 + 1
* +2
s +2
""
(s + 1) + 1
s* + 25 + 2

KlW ."
Krf,)

(7.15)

(7.16)

so that the inverse transforms are

Vl(t)

7.2

THEVENIN'S

= e- cos

/,

*(/)

- <r*(cos t + sin /)

(7.17)

AND NORTON'S THEOREMS

In network analysis, the objective of a problem is often to determine a

branch current through a given element or & single voltage across
an element. In problems of this kind, it is generally not practicable to
write a complete set of mesh or node equations and to solve a system of
equations for this one current or voltage. It is then convenient to use two

single

very important theorems on equivalent circuits,

and Norton's theorems.

known

as Thevenin's

Transform methods

in

network

181

anal/sis

-3X
Networks

n voltage

<S)

sources

Zi(s)

in current sources

FIG.

7.1

Thevenin's theorem
From the standpoint of determining the current I(s) through an element
can be
of impedance Z^s), shown in Fig. 7.1 1, the rest of the network
replaced by an equivalent impedance Ze(s) in series with an equivalent
voltage source Vs), as depicted in Fig. 7.12. The equivalent impedance
Zt(s) is the impedance "looking into" from the terminals of Zt(s) when
all voltage sources in JV are short circuited and all current sources are
open circuited. The equivalent voltage source V,(s) is the voltage which
appears between the terminals 1 and 2 in Fig. 7.1 1, when the element Z^s)

is

removed or open

theorem is

circuited.

that the elements in

The only requirement

for Thevenin's

Zx must not be magnetically coupled to any

element in N.

The proof follows. The network

current sources.

in Fig. 7.11 contains n voltage

and

We are to find the current I(s) through an element that

not magnetically coupled to the rest of the circuit, and whose impedance
1
According to the compensation theorem, we can replace Z^s)
x (s).
by a voltage source V(s), as shown in Fig. 7.13. Then by the superposition
principle, we can think of the current /(*) as the sum of two separate parts
is

is

Let the current It(s) be the current due to the n voltage and

<-

current

fl

VM

\2M

FIG.
1

7. IX

Thevenin's equivalent

Zi(e)

circuit.

Sons,
See H. H. Skilling, Electrical Engineering Circuits, 2nd Ed. John Wiley and

New York,

1965.

182

Network

analysis

and synthesis

we short circuit the source V(s), as shown in Fig. 7.14a.

equal to the short-circuit current 7gc Let IJs) be the
current due to the voltage source V(s) alone, with the rest of the voltage
sources short circuited and current sources open circuited (Fig. 7.146).
current and n voltage
With the
sources alone;

Therefore I^s)

i.e.,

is

sources removed in Fig. 7.146,

Network

N
it

voltage sources
current sources

FIG.

7.

that the network

)V(s)

7aC0

is

is

we

see

passive so that

related to the source V(s)

by the

relation

13

=-

V(s)
(7.19)

Zm(s)
the input impedance of the circuit at the terminals of the

where Zm(s)

is

source V(s).

We can now write /(*) in Eq.

/(s)

= Ms) -

Since Eq. 7.20 must be satisfied in

7.18 as

(7.20)

Zm(s)

all cases,

consider the particular case

when we open circuit the branch containing Zx(j). Then /($)

is the open-circuit voltage Kocfa). From Eq. 7.20 we have
he(s)

= Foc(s)
Z ln(s)

and

V(s)

(7.21)

Transform methods

in

network

analysis

183

FIG. 7.14b

so that

we can

rewrite Eq. 7.20 as

Zla(s)I(s) -

KocCs)

V(s)

(7.22)

In order to obtain the current I(s) through Zx (s), the rest of the network
can be replaced by an equivalent source Ve(s)
Voc(s) in series,
with an equivalent impedance Ze (s)
Zm(s), as shown in Fig. 7.12.

Example 7.4.

Let us determine by Thevenin's theorem the current I^s) flowing

through the capacitor in the network shown in Fig. 7.1S. First, let us obtain

LH0-)

'c<-)

FIG. 7.15

by opening all current sources and short circuiting

Then, we have in the network in Fig. 7.16, where

Z,(s)

Z,(s)

Next we find

Ve(s)

=R+sL

all

voltage sources.

(7.23)

by removing the capacitor so that the open-circuit voltage

1 and 2 is V,(s), as shown in Fig. 7.17.
We readily

between the terminals

184

Network

analysis

and synthesis

-rnnp-

>i

o2
FIG. 7.16

determine from Fig. 7.17 that

F,Cs)

-/(*)*+ Z/(0-)-

M0-)

(7.24)

By Thevenin's theorem we then have

/l(j)

VJto
~ Z.{s) + Z(s) =

7(j)Jt

+ Lt(0

-) -Pq(0-)/s
(7.25)

A+sL+ 1/sC

FIG. 7.17
7.5. For the network in Fig. 7.18, let us determine the voltage vjt)
0, and we
across the resistor by Thevenin's theorem. The switch closes at t
0.
assume that all initial conditions are zero at t

Fvampk

Lj

Li

=c

Qv(t)

itftf

FIG. 7.IS

terms of its transformed representation,

almost determine by inspection that the
can
which is given in Fig. 7.19. We
network to the left of AT in Fig. 7.19 is
the
of
source
voltage
Thevenin equivalent
First let

us redraw the

circuit in

VM

LiS

+ \jsC

(7.26)

Transform methods
Lis

La

in

network

analysis

185

* Vo

FIG. 7.19

and the input impedance to the

left

of AT is

slM/sC)
(7.27)

V&)

We know that
Therefore

V&)

"

ZJis)

RVisYMsC)
(R + sL^sLi + lIsC)

Finally

o(0

(7.28)

+R
+ LJC

(7.29)

(7-30)

e-HKtCi)]

Norton's theorem

When
tance

is

required to find the voltage across an element whose admitYifs), the rest of the network can be represented as an equivalent
it is

admittance Ys) in parallel with an equivalent current source I,(s), as

shown in Fig. 7.20. The admittance Yt(s) is the reciprocal of the Thevenin
impedance. The current I,(s) is that current which flows through a short
circuit across Y^s).

From

Fig. 7.20,

Vx{s) =
The element whose admittance

is

/.(*)

(7.31)

Yt(s) + YM
Yt must not be magnetically coupled to

any element in the rest of the network.

Consider the network in Fig. 7.21. Let us find the voltage across the
capacitor by Norton's theorem. First, the short-circuit current source
Ie (s) is found by placing a short circuit across the terminals 1 and 2 of the
Vi

91

()>

Yds)

FIG. 7.20

Yito

Network

186

anal/sis

and synthesis

-nppp>

Pi

.1
c

IM(

<P

'

FIG. 7.22

FIG. 7.21

capacitor, as

shown

From

in Fig. 7.22.

The admittance

Fig. 7.22 ls)

is

(7.32)

sL+R

Y,(s) is the reciprocal

'

Then

')

of the Thevenin impedance, or

(7.33)

sL+R

the voltage across the capacitor can be given as

V(s)

Example 7.6.
that v(t)

Yt(s) + Yds)

(sL

+ R)sC +

(7.34)
1

In the network in Fig. 7.23, the switch closes at / = 0. It is given

' and all initial currents and voltages are zero.
Let us find the

" O.le-6

/g(/) by Norton's theorem.

The transformed circuit is given in Fig. 7.24. To find the Norton equivalent
current source, we short circuit points 1 and 2 in the network shown. Then

current

Ie(s)

is

the current flowing in the short circuit, or

Us) =

0.1

V(s)

Rt +

The equivalent admittance of the

Y,(s)

J?!

circuit as

+sL

(s

5)

S)(s

viewed from points

+ sRxC +
R +sL

s*LC

- sC +

Us + RILXs +

sL

o-iTJoTP-wv

c=J=t

FIG. 7.23

0.5s*

(7.35)

10)

and 2

+ 5s +
* + 10

is

10
(7.36)

Transform methods

in

network

analysis

187

FIG. 7.24

I^s)

is

IM

then

+ GJ

(7.37)

+ 6)

(7.38)

/<(')

BJLYJto
1

(S

sy(s

By inspection, we see that /() can be written

+ 6) - (s + 5)
(s + 5)V +6)

(*
/(*)

as

(s

+ 5)*

(*

+ 6X* +

(7.39)
5)

Repeating this procedure, we then obtain

1

//*)

(*+5)

Taking the inverse transform of IJs), we

j+5 + j+6
finally

(7.40)

obtain

0 = ('"" - "" + e"*')(0

7.3

(7.41)

THE SYSTEM FUNCTION

As we

discussed earlier, a linear system

e(t) is related to

the response

tit)

by a

is

one in which the excitation

linear differential equation.

When

the Laplace transform is used in describing the system, the relation between
the excitation E(s) and the response R(s) is an algebraic one. In particular,

when we
related

discuss initially inert systems, the excitation

by the system function H(s) as given the

R(s)

We

how

= E(s)H(s)

and response are

relation
(7.42)

a system function is obtained for a given network,

can be used in determining the system response.
As mentioned in Chapter 1, the system function may assume many forms
and may have special names such as driving-point admittance, transfer
will discuss

and how

this function

188

Network

analysis

and synthesis

T
1

'R

'w(D

a?-r
FIG. 7.25

This is because the

impedance, voltage or current-ratio transfer function.
voltage
form of the system function depends on whether the excitation is a
or
current
specified
or current source, and whether the response is a
functions.
We now discuss some specific forms of system
voltage.

Impedance
a current source and the response is a voltage,
excitation and
then the system function is an impedance. When both
we have a
terminals,
pair
of
response are measured between the same
is
impedance
driving-point
impedance. An example of a

When

the excitation

is

driving-point

given in Fig. 7.25, where

m = m = R+
HKS)

I,(s)

I?22L

(7.43)

sL+l/sC

Admittance
a voltage source and the response is a current,
admittance IJV, is
H(s) is an admittance. In Fig. 7.26, the transfer
obtained from the network as

When

the excitation

is

H(s)=

m=

l_

(7.44)

Voltage-ratio transfer function

When the excitation is a voltage source and the response is also a voltage,
the voltagethen H(s) is a voltage-ratio transfer function. In Fig. 7.27,
ratio transfer function

V (s)lVt(s) is obtained as follows. We first find the

sL

innnp

*t

wvfit

1(

Si-

J2W

^=r
FIG. 7.26

Transform methods

in

ZiM

)yt

Z*8)

network

anal/sis

189

T
1

FIG. 7.27

current

Va(s)

I{s)sa

Z {s)+Zt(s)
VJs) = Z&)I(s)

(7.45)

Since

V (s)
Vg(s)

then

(7.46)

ZJs)

ZM + Z

(7.47)
t(s)

Current-ratio transfer function

the excitation is a current source and the response is another
current in the network, then H(s) is called a current-ratio transfer function.
As an example, let us find the ratio IJIa for the network given in Fig. 7.28.
Referring to the depicted network, we know that

When

I(s)

= IM + /(s),

Eliminating the variable Ilt

(7.48, 7.49)

sC

we find
(7.50)

so that the current-ratio transfer function

Jofr)

I(s)

1/sC

R + sL+1/sC

h(s)

7'

W Q)

is

4=ic

FIG. 7.28

(7.51)

Network

190

anal/sis

and synthesis

FIG. 7.29

From the preceding examples, we have seen that the system function is a
the
function of the elements of the network alone, and is obtained from
let
us
network by a straightforward application of Kirchhoff's laws. Now
system
the
obtain the response transform R(s), given the excitation and
the
function. Consider the network in Fig. 7.29, where the excitation is
that
assume
We
current source i(t) and the response is the voltage v(t).
0. Let us find
the network is initially inert when the switch is closed at t

the response V(s) for the excitations:

1-

i(t)

= (sin a>ot)u(i).

2.

i(0

is

3.

i,(0 has the

First,

we

the square pulse in Fig. 7.30.

waveform

in Fig. 7.31.

obtain the system function as

s

H(S)

sC
1.

it (t)

- (sin ou

1/sL

f) u(t).

+G

C[s

(7.52)

s(G/C)

The transform of i(t)

1/LC]

is

(7.53)

so that

V(s)

<Oo

= /(s)H(s) =
s*

a>

C[s*

s(GIC)

(7.54)

1/LC]

ig(t)
ig(t)

A
a

<i

FIG. 7.30

FIG. 7.31

Transform methods

For the square pulse

2.

in Fig. 7.30,

i(0

transform

Its

is

in

network

anal/sis

can be written as

ig {t)

= u(t) - u(t - a)
= i (i _

j,(s)

191

(7.55)

e-)

(7.56)

The response

V(s) is therefore given as

V(s)

+
C[s*

p-0*
-

s(G/C)

(7
57);
l

1/LC]

Note that in obtaining the inverse transform C-1 [F(j)], the factor e-*
must be regarded as only a delay factor in the time domain. Suppose we
rewrite V(s) in Eq. 7.57 as

K(s)-^(l-0

(7.58)

Then

if

we denote

the inverse transform by v^t),

we

r{^]

*<0

(7.59)

obtain the time response

v(t) <= Vl (t)

Observe that x(0 in Eq. 7.59

is

Vl (t

- a) u{t - a)

(7.60)

the response of the system to a unit step

excitation.

The waveform

3.

tf

and

V(s)

its

is

transform

then V(s)

in Fig. 7.31 can

we denote by

v,(t)

It (s)

(l

as

If

^^
=
+
+ 1 _ Cl)
+

(0

- (0

u(t

is

be represented as

a)

(7.61)

1 1
s \

as

as

3
as J C[s

(7.62)
/

s(GIQ

the response of the system to a

(7
K

63)

excitation,

we

1/LC]

ramp

'

see that

C-1 [F(s)]

= Vl (t) + - Vi(t) - - v (t a

where vt(t)

is

the step response in Eq. 7.59.

a) u(t

a)

(7.64)

192

Network

Let us

now

response R(s)

analysis and synthesis

discuss

some

further ramifications of the equation for the

= H(s) E(s).

Consider the partial-fraction expansion of

R(s):

R(s)

, v
!s s
+

By

A<
Si

(7.65)
t

where s( represent poles of H(s), and s, represent poles of E(s). Taking

the inverse transform of R(s), we obtain

rW^ZA^ + ZB^"*

(7.66)

The terms Af?** are associated with the system H(s) and are called free
response terms. The terms B^'* are due to the excitation and are known
as forced response terms. The frequencies st are the natural frequencies of
the system and j, arc the forced frequencies. It is seen from our discussion
of system stability in Chapter 5 that the natural frequencies of a passive
network have real parts which are zero or negative. In other words, if we
jo> { , then <x< <, 0.
<rt
denote st as st

initially inert network in Fig. 7.32, the excitation is

Let
us find the response v (t) and determine the free and
i cos t df).
forced response parts of y (r). The system function is

Example

For the

7.7.

va(t) =

Since V,{s)

K (*)

1/sC

VJis)

R +MsC

V,(s)

is

the response

is

2\s*

(7.67)

2
(7.68)

1/

then

W- VM =

0.4

(st

+ Ws + 2)

J+2 +

We next obtain the inverse transform

vjt) - -0.4* " + 0.4 cos t + 0.2 sin t

JK>

-2Q

-WW
C-Jf=i= uoW

FIG. 7.32

0.4*
*

+ 0.2
+l

(7.69)

(7.70)

Transform methods
It is

apparent that the free response

0.4 cos

As a

in

network

analysis

193

-0.4e~*, and the forced response

is

is

+0.2 sin/.

us consider the basis of operation

for the R-C differentiator and integrator shown in Figs. 7.33a and 7.336.
We use the Fourier transform in our analysis here, so that the system
function is given as H(ja>), where o> is the ordinary radian frequency
final topic in

variable.

Consider

our discussion,

first

let

the system function of the differentiator in Fig.

7.33c.

VJjw)

_ jaRC
jmRC +

_
R+

V,(J(o)

1/jcoC

(7.71)
1

Let us impose the condition that

coRC

(7.72)

We have, approximately,

*W =! jcoRC
Then

the response

(7.73)

VJjm) can be expressed as

(7.74)

Taking the inverse transform of

V^ijco),

we

obtain

nMsiRC-vJtl)

(7.75)

at

Note that the derivation of Eq. 7.75 depends upon the assumption that the
R-C time constant is much less than unity. This is a necessary condition.
Next, for the
function

R-C

integrator in Fig. 7.33ft, the voltage-ratio transfer

is

V/J*)
VJijm)

l/ja)C
1/jfoC

+R

jwCR

(7.76)
1

hgW

VO<t)

(a)

FIG. 7.33. (a)

(b)

R-C differentiator.

(6)

R-C integrator.

Network

94

If

we assume

analysis

that

and synthesis

o>RC )J>

1,

then

VM^T^zVjUa)
jcoRC
Under

these conditions, the inverse transform

so that the

7.4

(7.77)

is

(t)~--r Va(r)dT

(7.78)

RC Jo

R-C circuit in Fig.

7.336

is

approximately an integrating

circuit.

THE STEP AND IMPULSE RESPONSES

In this section we will show that the impulse response h(t) and the system

we can obtain step and

impulse responses directly from the system function.
We know, first of all, that the transform of a unit impulse (5(f) is unity,
i.e., C[<5(0]
1. Suppose the system excitation were a unit impulse, then
the response R(s) would be
function H(s) constitute a transform pair, so that

R(s)

We

= E(s) H(s) = H(s)

(7.79)

thus see that the impulse response h(t) and the system function H(s)

constitute a transform pair, that

is,

= H(S)
Z-^[H(s)] = h(t)
Z[h(t)]

(7.80)

Since the system function is usually easy to obtain, it is apparent that we

can find the impulse response of a system by taking the inverse transform
of H{s).

Example

7.8.

Let us find the impulse response of the current

The system function is

/(/) in

the

R-C

circuit in Fig. 7.34.

~
Simplifying H(s) further,

VM ~ R + 1/*C ~ R(s + 1/RO

we have

^^O-JTw)
The impulse response
Ht)

which

is

shown

is

(7 ' 81)

(7 - 82>

then
C-i[/f(5)]

in Fig. 7.35.

= i ^r) - -L e-wj

(,)

(7.83)

Transform methods

network

in

analysis

195

i(t)

)9(t)

/Z-t/RC
_

R*C
FIG. 7.34

FIG. 7.35. Current impulse

R-C network in

response of
Fig. 7.34.

Since the step response is the integral of the impulse response, we can
use the integral property of Laplace transforms to obtain the step response
as

W-rf?]
where

oc(f)

denotes the step response.

Similarly,

(7.84)

we

obtain the unit

ramp

response from the equation

^-m

(7.85)

where y(t) denotes the ramp response. From this discussion, it is clear
that a knowledge of the system function provides sufficient information to
obtain

all

the transient response data that are needed to characterize

the system.

Example
7.36.

7.9.

Let us find the current step response of the R-L circuit in Fig.
is the response and V (s) is the excitation, the system function
g

Since I(s)

is

V{s)

Therefore H(s)/s

is

h{s)
s

The

(7.86)

R+sL

step response </) is

n_ i \
=i
s(R+sL) R\s
s + RjLJ

now

(7.87)

obtained as
1

a(/)=-(l -e-W*X)ii(/)

To

check

this result, let us consider the impulse response of the

(7.88)

R-L

circuit,

Network

196

which we found

analysis

in

and synthesis

Chapter

5.
1

(7.89)

Kt)

The

step response

the integral of the impulse response, or

is

a(0

</t=-(i -*-w*>0(i)

(7.90)

-J>

if we know the impulse response of an initially inert

the response of the system due to any other
determine
can
linear system, we
excitation. In other words, the impulse response alone is sufficient to characterize the system from the standpoint of excitation and response.

It is readily

seen that

R
-VSMAK)

V,M{

FIG. 7.37

FIG. 7.36

system
7.10. In Fig. 7.37, the only information we possess about the
in the black box is: (1) it is an initially inert linear system; (2) when v&) = 6(t),

Example
then

With this information, let us determine what the excitation vjf) must be in order
to produce a response vf)

- te~" u(f).

First,

we determine the system function

to be

H(s)

V^s)

Vjs)

7T2

2*4-5

t j+3

We next find the transform of te~* (*)

Vjs) - titer"] excitation is then

VAs)
Simplifying

2(a

(7.92)

+3)

(7.93)

+2)*

found from the equation

JW "
H(s)

and expanding V&s)

K<W =
The system

+ 2X

<

The unknown

(j

(s

+ 2Ks + 3)
2.5X* + 2)

(7.94)

20 +

into partial fractions,

1
(s + 3)
+ 2.5X + 2) " s + 2

we have
0.5
*

+ 2.5

(7.95)

excitation is then

vtf)-{e-*t -0.5e-*

M)u(t)

(7.96)

Transform methods

7.5

in

network

analysis

197

THE CONVOLUTION INTEGRAL

In this section we will explore some further ramifications of the use of

the impulse response Mj) to determine the system response r(t). Our
discussion is based upon the important convolution theorem of Laplace
(or Fourier) transforms. Given two functions /,(*) and//*), which are
zero for t
0, the convolution theorem states that if the transform of

x (f) is

<
Ft(s),

and

if

convolution of/i(f)

WTO,

that

the transform of fi)

and/t(r)

is

Fs), the transform of the

is,

- t)/,(t) dr] = F1(s) F&)

t[ j*Mt

where the

is

the product of the individual transforms,

is

integral

ft(t

(7.97)

t)/,(t) dr

the convolution integral or folding integral,

and

is

denoted operationally

as

fm -

Let us prove that C[/, */J

Proof.

C|/i(0V^0]

From

t)/^t) dr

dt

(7.99)

the definition of the shifted step function

- t) =

=
we have the

Then Eq.

- t)/,(t) dr =

r>t

(7.100)

f 7i(

" t) (* - t)A(t) dr

(7.101)

Jo

7.99 can be written as

C[/i(0Vi(0]

= f "e- f 7i(* - t) (t - r)Mr) dr dt

(7.102)

Jo

Jo

let

rt

identity

f/i('
Jo

we

(798)

= F&. We begin by writing

= V"[ J/iO ~ r)Mr) dr]

u(t

If

- fflVMt)

= r so that
t

'

-.t

e-^f,>

(7 103)

Network

198

analysis

and synthesis

then Eq. 7.102 becomes

[/i(0* /*(')]

I*" f"'f1 (x)u(x)fi (T)e-'e->*dTdz

Jo Jo

f 7i()
Jo

u(x)e-" dx f 7(r)e- w dr
Jo

= F (s)F2(s)

(7.104)

The separation of the double integral in Eq. 7.104 into a product of two
integrals is based upon a property of integrals known as the separability
property. 2
-2

Example 7.11. Let us evaluate the convolution of the functions/i(f) = e ' u(t)
and fift) = t u(t), and then compare the result with the inverse transform of
F^s) FJs), where

(7.105)

W
The convolution
variable

of_/i(f)

t for

and/2(f)

p
obtained by

is

first

substituting the

dummy

in/i(/), so that

fiff

T)

c-* (

'-T)

u(t

(7.106)

t)

Then/iWVaOis

f *fi(f

T)/*( T)

rfT

Integrating

let

^T = ~a*

T 2r

dr

(7.107)

Jo

by parts, we obtain
fi(0* fiff)

Next

_2
r " _T)
I

Jo

Jo

({-\+\ *-*) W

us evaluate the inverse transform of F^s) Fs).

(7-108)

From Eq.

7.105,

we

have

C-Vito *a(*)l = (^ - ^ + \ e~*)

so that

An

important property of the convolution integral

equation

.
t

flit

See, for example, P. Franklin,

New

York, 1940.

Jo

Treatise

is

71 10)

expressed by the

- r)/a(r) dr =

Jo

Sons,

"(')

MMA -

r)

dr

(7.111)

on Advanced Calculus, John Wiley and

Transform methods
This

is

readily seen

from the relationships

network

in

= *i(')
CWOViC)] - Wi()
CC/i(0V*]

and

199

analysis

(7-112)

(7113)

To give the convolution integral a more intuitive meaning, let us examine

the convolution or folding process

from a graphical standpoint. Suppose

we

JA(t)

Mr)

= t (t)

take the functions

shown

as

- ,

and

in Figs. 7.38a

obtaining the integral

7.39a.

m(t)

(7.114)

In Fig. 7.38, the various steps for

.,

A('-t)/2(t)V
Jo

= ( t).

are depicted. Part (b) of the figure shows /i( t)

ft(t

in part (c) merely advances

product

A(|
is

shown in part (d).

- u(t - t)

t)

/x (t)
.

t)/

(t)

The

(7.115)

by a variable amount

t.

_ t)t u(t)

<f

function

Next, the
(?

n6)

We see that the convolution integral is the area under

the curve, as indicated by the cross-hatched area in part

convolution integral has a variable upper limit,

under the curve of/i(f
T)ft(r) for all t. With
Fig. 7.38a", the area under the curve is

AV.-/(0 =

we must
t

(</).

Since the

obtain the area

considered a variable in

(7.117)

!J(0

as plotted in part (e) of the figure.

In Fig. 7.39,

same

we

result as in

taking the inverse

by folding g(r) about a point t, we obtain the

Fig. 7.38. The result in Eq. 7.117 can be checked by
transform of F^s) Fjfa), which is seen to be
see that

c- 1 [F1 (s)fW]

-1

l3j

m(o

(7118)

Let us next examine the role of the convolution integral in system

From the familiar equation

analysis.

R(s)

we

= E(s) H(s)

(7.119)

obtain the time response as

r(t)

where e(r)

is

"*[() H(s)]

the excitation

and

('e(r) h(t

- t) dr

(7.120)

k(r) is the impulse response of the system.

200

Network

analysis

and synthesis

Mr)

Mr)
1

/
/

(a)

(a)

\M-t)

M-r)
1

T
(b)

M-t)

Mt-r)
1
1

(e)

hM M-t}

M-r) Mr)

id)

ffh

M
FIG. 7.39

FIG. 7.38

Using Eq. 7.120, we obtain the response of a system directly in the time
domain. The only information we need about the system is its impulse
response.
EVro-pW. 7.12a.

Let us find the response /(f) of the R-L network in Fig. 7.40 due

to the excitation

c(t)

- 2<r* y(f)

The impulse response for the current

is

.AW-|*-raBW
Therefore, for the

R-L

circuit

(7.121)

(7.122)

under discussion

Kt) *-"(*)

(7-123)

Transform methods

in

network

Using the convolution integral, we obtain the response

Kt)

V~

t)

Kr) dr

i(i)

201

analysis

as

- 2j V"-rtr* </r
(7.124)

- 2-< f V"*</t - 2(-* - *-*)(0

VvWvV-

0*"

FIG. 7.41

PIG. 7.40

The ideal amplifier in Fig. 7.41 has a system function H(s) K,

s 7.12b.
where Kis a constant. The impulse response of the ideal amplifier is then

Kt)-Kd(t)

(7.125)

Let us show by means of the convolution integral that the response r{t)
to the excitation e(t) by the equation

K0-JT4)
Using the convolution

K0

integral,

is

related

(7.126)

we have

(t)*('-t)/t-a:

r)dr-Ke(f)

(7.127)

r-

a variable in the expression for r(t), we see that the ideal amplifier in
is an impulse-scanning device which scans the input (/) from
domain
the time
/
to f oo. Thus, the response of an ideal amplifier is a replica of the
input e(f) multiplied by ibcgain IT of the amplifier.
Since

7.

/ is

THE DUHAMEL SUPERPOSITION INTEGRAL

In the Section 7.5, we discussed the role that the impulse response plays
in determining the response of a system to an arbitrary excitation. In
this section we will study the Duhamel superposition integral, which also

describes

an input-output

relationship for a system.

The superposition

integral requires the step response a(f) to characterize the system behavior.
plan to derive the superposition integral in two different ways.

We

Network

202

The

simplest

is

analysis

and s/nthesis

examined

We

first.

relationship

R(s)

begin with the excitation-response

= E(s) H(s)

(7.128)

Multiplying and dividing by s gives

R(s)

E(s)

(7.129)

Taking inverse transforms of both sides gives

C- 1 [(5)]

C-1

[^- -s(s)]
(7.130)

= -ip)l* t-i [s(s)]

which then

yields

K0 =

a(0*[e'(0

+ *(0-)d(0]

= e(0-) a(f) +
where

and

e'(t) is

<x.(t)

is

the derivative of

e'(r) a(f

e(0 )

7.13.

Duhamel superposition

Let us find the current

the voltage source

- t) dr

(7.131)

the value of e(i) at

is

Equation 7.131

the step response of the system.

referred to as the

Example

e(t),

'

J
/

is

0,

usually

integral.

i(t) in

the

R-C circuit

in Fig. 7.42

when

is

(7.132)

as

shown

The system function of the R-C circuit

in Fig. 7.43,

H(s)

Us)

V(s)

R(s

Therefore the transform of the step response

M)
s

is

(7.133)

IjRC)

is

R(s

(7.134)

l/RC)
vg (t)

*WI

Slope

a A2

zizc

FIG. 7.42

FIG. 7.43

Transform methods

in

network

analysis

203

Taking the inverse transform of H(s)/s, we obtain the step response

1

~-tlRC

<t)

From the excitation in Eq.

is

see that e(0

(7.135)

) =0 and

= A 1 d(f) + A t u{t)

v'^i)

The response

we

7.132,

(0

(7.136)

then
i(0

-jTj'V,(r)a(t -r)dr

A x Hr)(t
<5(t) o(/ -

t) dr
r)dr

At <x(f -

r)dr

Jo-

Jo-

e-Ur-r)IRC fc

ic

/{

(7.137)

Jo_

- [J e-"* + /<* C(l - e-**

)] u(t)

from the Example 7.13, e(0 ) = 0, which is the case in many

Another method of deriving the Duhamel integral
avoids the problem of discontinuities at the origin by assuming the lower
limit of integration to be t = 0+. Consider the excitation e(t) shown by
the dotted curve in Fig. 7.44. Let us approximate e(t) by a series of step

As we

see

transient problems.

functions, as indicated in the figure.

We can write the

staircase approxi-

mation of e(0 as
e(t)

= e(0+) h(0 + &E1 u(t- At)

+ AE, - 2At) +
i/(f

+ An u(t - /iAt)

'11
Signal

}a3

s
/

-A2

hABi

<P+>

0h-AT-|

2At

_L
3At

4At

FIG. 7.44. Staircase approximation to a

signal.

(7.138)

Network

204

analysis

and synthesis

where AE^ is the height of the step increment at

*At. Since we assume

the system to be linear and time invariant, we know that if the response to a
t). Therer) is X< a(f
unit step is <x(f), the response to a step K+ u(t
fore we can write the response to the step approximation in Fig. 7.44 as

K0 =
If

At is

K0 -

e(0+) a(0

small, r(t)

+ AEX a(/ - At)

+ A^a(r 2At) +

+ Aa(f - nAf)

(7.139)

can be given as

e(0+) o(0

At

AE

* oc(*

- At) At
(7.140)

+ *h (, - 2At)At +

+ ^aO -

At)At

At

At

which, in the limit, becomes

AT

n
r(t)

(0+) (0

lim 2
+ Ar-0<-+

~
tV
AT

'

At) At

1-CO

(7.141)

= e(0+) a(0 + f e'(r) ( - t) dr

Problems
In the circuit shown
sketch^;)7.1

v(t)

.- 2(f)

and /(0-)

-2

amps.

Find and

100
r

VWfe|

vwQ

1000

5h

PROB.7.1
after having been at
7.2 The switch is thrown from position 1 to 2 at t
t
1e~' sin fit.
for a long time. The source voltage is vjif)

(a)

Find the transform of the output voltage

(b)

Find the

initial

vtf).

and final values of vjt)-

plane
Sketch one possible set of locations for the critical frequencies in the *
transform.)
and write tint form of the response v&). (Do not take the inverse

(c)

Transform methods

in

network

anal/sis

205

MtOB. 7.2

13
/(f)

for

In the circuit shown, all initial currents and voltages are zero.
using Thevenin's theorem.
t >

Find

51(0

PROS. 7.3

7.4 In the circuit shown in (a), the excitation is the voltage source e{t)
described in (6). Determine the response i(t) assuming zero initial conditions.

e(t)

10

*t)

'S*

4=i*

1/2

-1

M
PROS. 7.4

13

Determine the expression for vji) when /(/)

Use transform methods.

conditions.

Hi), assuming zero initial

Network

206

anal/sis

and synthesis

PROB. 7.5
7.6

The

opened.
0.5 sin

circuit

shown has zero

Find the value of the

Vlt

n(0.

The excitation

initial

resistor

is i(t)

energy. At t =
the switch S
that the response is v(t)

X such

te~ **

is

u(t).

i
v(t)

=r*

PROB. 7.6
7.7 Use transform methods to determine the expressions for i^t) and /j(/)
-10 '.
Assume zero initial
lOOe
in the circuit shown. The excitation is v(f)
energy.

1000

K=\

PROB. 7.7
circuit shown, the switch S is opened at / = 0. Use Thevenin's
to determine the output voltage t^f). Assume zero initial
theorem
or Norton's

7.8

energy.

For the

Transform methods

in

network

207

analysis

PROB. 7.8

7.9

6u(0>

For the transformer shown, find i^t) and ,(/). It is given that e(t) =
and that prior to the switching action all initial energy was zero; also

M=1A.

PROB. 7.9

7.10

For the circuit shown, find /,(/), given that the circuit had been in steady

state prior to the switch closing at

10Q

WWW

0.

Kml

100

WW
400

vl=
2hojlo8h

j=z=_40v

PROB. 7.10

7.11

cos 20

Find
u(t).

/,(/)

using Thevenin's theorem.

Assume zero

initial

energy.

The

excitation is e(t)

100

Network

208

anal/sis and synthesis

TVTP-

y/W

2h

10O

300
7.50

|6h

PROB. 7.11
7.12
(/),

Determine the transfer function H(s)

find v t{i).

Assume zero

initial

= V^IV^s). When

v^t)

conditions.

10

r-nAAA-

if

=j=

T
v
*

G) 5V

=r

3f

fOS v3

PROB. 7.12
7.13
find v(t)

Using (a) standard transform methods and

when

/(/)

(b) the

convolution integral,

2e~* u(t). Assume zero initial energy.

mo

m
lf=t

io:

ih<

PROB. 7.13
7.14

linear system is shown in the figure. If the

determine the response values r(l) and HA,)

The impulse response of a

excitation were

<*/) * 3e"*'(0,

using graphical convolution.

PROB. 7.14

Transform methods

in

network

analysts

209

is given as H(s) = l/(s* + 9)*. If the excitation

determine the response #(') (Hint: Use the convolution
integral to break up the response transform R(s).)

7.15

were

The system function

c(/)

7.16

= 3S'(t),

Solve the following integral equations for

(a)

(0

(*)

sinr

f(f

t)x(t)/t

x(t).

a<T)e-<-f >/T
J

(/)

7.17

+j

Using the convolution

x(jt

r) e-*

dr

=* t

integral find the inverse transform of

K
()

(s

+ aXs + b)

2fr+2)
(s*

+ 4)

(s*

F(s)

<<0

7.18

"

By

1)

graphical means, determine the convolution of /(/)

(i.e., determine/(/)*/(/)).

shown

in the

figure with itself

T
PROB.7.18
7.19

Using

(a) the convolution integral

integral, find v(t) for e(t)

and

(b) the

Duhamel

superposition

= 4e-**u(t). Assume zero initial conditions.

PROB.7.19

210
7.20

Network
Using

integral, find

analysis

(a) the

<0 for

and synthesis

convolution integral and (A) the Duhamel superposition

= 2e-s * (/). Assume zero initial conditions.

e(t)

PRO B. 7JO

K(i)//(*) has the

7.21 For the circuit in (a) f the system function H(s)
and C. If the excitation '(/)
poles shown in (c). Find the element values for
has the form shown in (b), use the convolution integral to find v(t).

(b)

C-0.5 ju

-r

-2

-1

to

PROS. 7.21
7.22 The excitation of a linear system is x(r), shown in (a). The system
impulse response is Hf), shown in (ft). Sketch the system response to x(t).
neat, carefully dimensioned sketch will
(No equations need be written.

suffice.)

Transform methods

in

network

anal/sis

21

(a)
h(t)

(b)

PROS.
7.23
is /(/)

7.22

A unit step of voltage is applied to the network and the resulting current

= 0.01*^' + 0.02 amps.

(a)

Determine the admittance Y(s) for

this network.

(b)

Find a network that

admittance function.

will yield this

PROS. 7.23

PROB.

7.24

7.24 The current generator delivers a constant current of 4 amps. At

the switch 5 is opened and the resulting voltage across the terminals 1, 2 is

c(r)

(a)
(b)

=6e-'

12v.

Find Z(s) looking into terminals 1, 2.

Find a network realization for Z(s).

chapter

Amplitude, phase, and delay

8.1

AMPLITUDE AND PHASE RESPONSE

In this section we will study the relationship between the poles and
zeros of a system function and its steady-state sinusoidal response. In
other words, we will investigate the effect of pole and zero positions upon
the behavior of H(s) along the jw axis. The steady-state response of a

system function

is

given by the equation

H(j<o)

where

=M {my*

(8.1)

an
or magnitude response function, and
odd
and
is
an
^(o>) represents the phase response,
is

Af(a>) is the amplitude

even function in o>.

function of to.
The amplitude and phase response of a system provides valuable
information in the analysis and design of transmission circuits. Consider
the amplitude and phase characteristics of a low-pass filter shown in
Figs. 8.1a

and

8.1 A.

The

cutoff frequency of the filter is indicated on the

c. It is generally taken to be the "half-

amplitude response curves as

power" frequency at which the system function \H(jta)\

is

equal to 0.707

amplitude \H(j(Omu)\. In terms of decibels, the halfpower point is that frequency at which 20 log \H(j(o c)\ is down 3 db from
20 log |#(/a)max)|. The system described by the amplitude and phase
characteristics in Fig. 8.1 shows that the system will not "pass" frequencies

of the

maximum

Suppose we consider a pulse train whose

We know
amplitude spectrum contains significant harmonics above co
block all
will
but
m
below
harmonics
that the system will pass the
c,
distorted
will
be
train
pulse
output
the
harmonics above (oc Therefore
harmonic
higher
many
because
train,
pulse
original
when compared to the
terms will be missing. It will be shown in Chapter 13 that if the phase

that are greater than <o a .

212

Amplitude, phase, and delay

213

Amplitude
response

0707

-b

fc
(a)

Phase response

FIG.

response

8.1.

Amplitude and phase response of low-pass

then

<f>{w) is linear,

from the phase response

filter.

minimum pulse distortion will result.

in Fig. 8.1 that the phase

<f>(a>)

is

We see

approxi-

mately linear over the range (o c <,w<, +co c If all the significant
harmonic terms are less than coc then the system will produce
minimum phase distortion. With this
example, we see the importance of an
.

-VWVAr

amplitude-phase description of a system.

In the remaining part of this chapter,

we

concentrate

will

on methods to

ViW

obtain amplitude and phase response

curves,

=r

V2 (s)

both analytically and graph-

ically.

To
curves,

obtain

we

let

amplitude
s

=ju>

and

phase

FIG.

84

in the system

and express H(ja>) in polar form. For example, for the amplitude
and phase response of the voltage ratio VJVX of the R-C network shown

function,

in Fig. 8.2, the system function

H(s)

is

V^s)
Letting

ljRC

F^s)

= jot, we see that H(Jco)

H(Jm)

l/RC

(8.2)

is

l/RC

jm

l/RC

(8.3)

214

Network

analysis

and synthesis

M(w)

FIG.

8.3.

Amplitude and phase response of R-C network.

In polar form HQco) becomes

HO) = (to

l/RC
2

1/K

--/tan-i

C2)*

a>BC_
=

M(ft>)e'*

(0,)

(8.4)

The amplitude and phase curves are plotted in Fig. 8.3. At the point
(o = 0, the amplitude is unity and the phase is zero degrees. As a> increases,

When co *= l/RC, the

45. This point is the half-power
Finally as <o - oo, Af(o>) approaches

the amplitude and phase decrease monotonically.

amplitude

is

0.707 and the phase

is

point of the amplitude response.

zero and ^(g>) approaches 90.

Now let us turn to a method to obtain the amplitude and phase response

from the pole-zero diagram of a system function. Suppose we have the

system function
H(s)

Ms - Zq)(s -

=
(s

)(s

Pi)(

Zi)

- Ps)

H(jco) can be written as

HUco)

*,)0
*P
Zq)0'<>gp
0>
Pi)(J> - P*)
Po)0>
jco z or jco p corresponds

(8.5)

ApO'cu

(g6)

to a vector
Each one of the factors
t
}
from the zero zi or pole/*, directed to any point jco on the imaginary axis.

Therefore,

if

we

express the factors in polar form,

jco

- z, =

rV",

jco- p,=

M,e'*'

(8.7)

i(y><>+v>i-a-i-ti)

(g 8 )

then H(j(o) can be given as

jj(jca)

as

shown

in Fig. 8.4,

j4

NNl

MoMiMj

where we note that

0! is negative.

Amplitude, phase, and delay

FIG.

8.4.

In general,

215

Evaluation of amplitude and phase from pole-zero diagram.

we can express the amplitude response

M(co) in terms of the

following equation.
n

H
M(.) = ^-

vector magnitudes from the zeros to the point

XI vector magnitudes from

Similarly, the phase response

is

on

thejco axis

the poles to the point on the/co axis

given as

^(eu)

=J

angles of the vectors from the zeros to the/co axis

2
It is

angles of the vectors from the poles to thejco axis

important to note that these relationships for amplitude and phase

are point-by-point relationships only. In other words, we must draw

vectors from the poles and zeros to every point on thejco axis for which we

wish to determine amplitude and phase. Consider the following example.

F(s)

=5*

4s

4s

+ 2s + 2

(s

+ H-jl)(s+l-jl)

(8.9)

Let us find the amplitude and phase for FQ2). From the poles and zeros
2, as shown in Fig. 8.5. From
of F(s), we draw vectors to the point to

216

Network

anal/sis

and synthesis

m
J2

V5/

/\45*
x

>

Vio7

>

-l

/\71.8*
x

FIG.

8.5.

-n

Evaluation of amplitude and phase from pole-zero diagram,

the pole-zero diagram,

it is

MQ2)
and

= 4(

,_

,- )

= 90 - 45 - 71.8 =

<f>(j2)

With the values M(j2) and

four additional points,

clear that

<f>Q2)

1.78

-26.8

and the amplitude and phase

we have enough

at three or

information for a rough estimate

JU
X

\M5'

yi

XX90*

-s.90*

1.0

'

/S
1A5

-yi

(a)

FIG.

8.6.

Br.

-j

(6)

Determining amplitude and phase at zero and very high frequencies.

Amplitude, phase, and delay

217

=
=

of the amplitude and phase response. At w

0, we see that the vector
magnitude from the zero at the origin to to
0, is of course, zero.
0. From Eq. 8.9 for F(s), F(j0) is
Consequently, M(j0)

lim F(ja>)

4Q0)

(8.10)

(l+Jlxl-Jl)

o>>0

this equation, we see that the zero at the origin still contributes a
90 phase shift even though the vector magnitude is zero. From Fig.
45
45
90.
90
is #0)
8.6a we see that the net phase at <w

From

approximately equal to

coJfii,0,

where

Extending

y>

all

1,

the vectors are

Then

~ 4eoh _
ca k

#a>)

toh

as seen in Fig. 8.66.

M(w)

and

mh

Next, at a very high frequency

<

~ 90 - 90 - 90 =

-90

this analysis for the frequencies listed in

Table

8.1,

we

obtain

and phase as given in the table. From this table we

sketch
the
amplitude
and phase curves shown in Fig. 8.7.
can
Next let us examine the effect of poles and zeros on the jto axis upon

values for amplitude

frequency response. Consider the function

F(s)

= s* +
s* +

1.03
1.23

_ (s + jl.015)(s - jlMS)
~ (s+ ;1.109Xs - jl.109)

(8.11)

Amplitude

10

+90

0123456789

-90

Frequency,

FIG. C7. Amplitude and phase response for F(s)

in

Eq.

8.9.

10

218

Network

analysis and synthesis

TABLE
Frequency,

8.1

Amplitude

CO

Phase, degrees

1.0

1.8

25.8

1.5

2.0

-5.3
-50.0
-66.0
-78.5

3.0

1.3

5.0

0.8

10.0

0.4

whose pole-zero diagram is shown in Fig. 8.8. At co <= 1.015, the vector
from zero to that frequency is of zero magnitude. Therefore at a zero
on the /o> axis, the amplitude response is zero. At co = 1.109 the vector
from the pole to that frequency is of zero magnitude. The amplitude
response is therefore infinite at a pole as seen from Eq. 8.11. Next,
consider

<

the

phase response. When

apparent from the pole-

JU

co

y'1.109

zero plot that the phase

co

<J./1.015

>

1.015,

1.015

it is

and o><

from the zero

at

co

is

When

zero.

1.109, the vector

1.015

is

now

pointing upward, while the vectors from

the other poles and zeros are oriented

O-./1.015
:

in the

same

direction as for co

We see that at a zero on the yw

-y'1.109

<

1.015.

axis, the

phase response has a step discontinuity

of +180 for increasing frequency.
Similarly, at a pole on the jco axis, the
phase response is discontinuous by 180. These observations are
illustrated by the amplitude and phase plot for F(s) in Eq. 8.11 for the
frequency range 0.9 <; co ^ 1.3, shown in Fig. 8.9.
With a simple extension of these ideas, we see that if we have a zero at
z a jco t , where a is very small as compared to co t , then we will have
a dip in the amplitude characteristic and a rapid change of phase near
co = co t as seen in Fig. 8.10. Similarly, if there is a pole atp = a jco t
with a very small, then the amplitude will be peaked and the phase will
decrease rapidly near co
co t as seen in Fig. 8.11. A contrasting situation
occurs when we have poles and zeros far away from they'd) axis, i.e., a is
large when compared to the frequency range of interest. Then we see that
these poles and zeros contribute little to the shaping of the amplitude and
phase response curves. Their only effect is to scale up or down the overall
amplitude response.
From stability considerations we know that there must be no poles in
the right half of the s plane. However, transfer functions may have zeros
FIG. 8.8

Amplitude, phase, and delay

1.015

FIG.

8.9.

in the right-half plane.

and

8.126.

219

1.109

Amplitude and phase far F(s) in Fig.

8.8.

Consider the pole-zero diagrams in Figs. 8.12a

Both pole-zero configurations have the same poles; the only

difference is that the zeros in (a) are in the left-half plane at s

while the zeros in (b) are the mirror images of the zeros in

(a),

]1,

and are

= +1 jl. Observe that the amplitude responses of the

two configurations are the same because the lengths of the vectors correspond for both situations. We see that the absolute magnitude of the
located at s

5.

<E

NX"

FIG. 8.10. Effect of zero very near

FIG.

the/co axis.

the/o> axis.

8.1 1. Effect

of pole very near

220

Network

analysis

and synthesis
JW

J2

>2

yi

-n
X

-y'2

-J2

(a)

(a)

<r

o-;i -

FIG. 8.12.

-2

-l

fl

-2

(b)

Minimum phase function,

(b)

Nonminimum phase

function.

phase of (b) is greater than, the phase of (a) for all frequencies. This is
because the zeros in the right-half plane contribute more phase shift
(on an absolute magnitude basis) than their counterparts in the left-half
system
plane. From this reasoning, we have the following definitions.
function with zeros in the left-half plane, or on the jco axis only, is called a
minimum phase function. If the function has one or more zeros in the

a nonminimum phase function. In Fig. 8.13, we see

the phase responses of the minimum and nonminimum phase functions
right-half plane,

in Figs. 8.12a

it is

and

8.126.

Let us next consider the pole-zero diagram in Fig. 8.14. Observe that
the zeros in the right-half plane are mirror images of the poles in the
100
y- Minim urn phase

l-ioo

1
^^/" Nonminimum phase
YY)
-360
10

12

FIG. 8.13. Comparison of minimum and nonminimum phase functions.

Amplitude, phase, and delay

left-half plane.

Consequently, the vector

drawn from a pole to any point

co x

JO>

on

>2

they a> axis is identical in magnitude with

the vector drawn from its mirror image

apparent that the amplitude

response must be constant for all frequencies. The phase response, however, is
anything but constant, as seen from the
amplitude and phase response curves
to

<!.

221

Kml
~

It is

given in Fig. 8. 1 5 for the pole-zero config-

-2 -i

x-yi

FIG. 8.14. All-pass function.

uration in Fig. 8.14.

A system function whose poles are only in the left-half plane and whose
zeros are mirror images of the poles about the jm axis
function.

The networks which have

is

called

an

all-pass

all-pass response characteristics are

often used to correct for phase distortion in a transmission system.

8.2

BODE PLOTS

In this section we will turn our attention to semilogarithmic plots of

amplitude and phase versus frequency. These plots are commonly known
as

Bode plots. Consider the system function

H(s)

Ms)

(8.12)

D(s)

1.0

10

*-

+ 180

l -I*
-360
10

FIG. 8.15. Amplitude and phase of all-pass function in Fig.

12

8.14.

Network

222

analysis

and synthesis

Ka

>l

3
S

togu-

K2

l*KI
Magnitude
(a)

K positive
Phase

K negative

e-

IT
logw(b)

FIG.

We know that the

Magnitude and phase of constant.

8. 16.

amplitude response

M()

is

|HC/)I

- j^}

(8.13)

\D(J>)\

If

we

express the amplitude in terms of decibels,

20 log M(w)

= 20 log

\N(jco)\

In factored form both N(s) and D(s) are

- 20 log

D(jco)\

(8.14)

made up of four kinds of factors:

(a)

a constant,

(b)

a root at the origin, s

(c)

a simple real root, s

(d) a

we have

complex pair of roots, s2

+ 2<w +

a*

/?*

To understand the nature of log-amplitude plots, we need only examine the

amplitude response of the four kinds of factors just cited. If these factors
are in the numerator, their magnitudes in decibels carry positive signs.
If these factors belong to the denominator, their magnitudes in decibels
carry negative signs. Let us begin with case (a).
(a)

The factor K. For the constant K, the db loss (or gain)

20 log

K = Kt

is

(8.15)

Amplitude, phase, and delay

223

t is either negative if \K\ < 1, or positive if \K\ > 1. The

phase response is either zero or 180 depending on whether is positive or
negative. The Bode plots showing the magnitude and phase of a constant

The constant

are given in Fig. 8.16.

The factor

(b)

The

s.

loss (gain) in decibels associated with

20 log

(zero) at the origin is

Thus the plot of magnitude

to.

versus frequency in semilog coordinates

20

db/decade or

is

constant for

For convenience,

a.

2Qlog

shown

shown

+^=

The phase

in Fig. 8.18a.

Arg
as

is

Bode

plots in Fig. 8.17,

at to

1,

we

and the phase

is

all co.

(c) The factor s

magnitude is

as

a straight line with slope of

6 db/octave. From the

see that the- zero loss point (in decibels)

a pole

in decibels

O+

let

us set a

2Qlog(o)S

1.

Then the

, )M

(g

lfi)

is

l)*1

tan~x

a>

(8.17)

in Fig. 8.186.

A straight-line approximation of the actual magnitude versus frequency

curve can be obtained from examining the asymptotic behavior of the
factor y<o
1. For to
1, the low-frequency asymptote is

20 log

For

to

1,

\j(o

a<1 s* 20 log

the high-frequency asymptote

201og|7o

0.1

1|

02

03

0.5

1.0

Frequency,

FIG.

8. 17.

db

(8.18)

is

20 log

l|

0.7

m db

2.0

3.0

(8.19)

5.0

Magnitude and phase of pole or zero at s

= 0,

7.0

10.0

224

Network

anal/sis

and synthesis

12

(+!)->.

2
8"

025

0.5
(ii

1
fc-

(b)

FIG. .!. Magnitude and phase of simple real pole or zero.

Amplitude, phase, and delay

TABLE

a> =-

Approximation,

Error,

db

db

db

2 octaves below
octave below
break frequency
octave above
2 octaves above

=i
o) = 1
m =2

o)

o-4

8.2

Actual
Magnitude,

Frequency

225

Straight-Line

0.3

0.3

1
3

12.3

12

1
0.3

we saw in (b), has a slope of 20 db/decade or 6 db/octave. The

1, which we designate
low- and high-frequency asymptotes meet at m
as the breakfrequency or cutofffrequency. The straight-line approximation
is shown by the dashed lines in Fig. 8. 1 8a. Table 8.2 shows the comparison
between the actual magnitude versus the straight-line approximation. We
which, as

see that the

maximum

error

is

at the break frequency

co

1,

or in un-

a.
normalized form: m
For quick estimates of magnitude response, the straight-line approximation is an invaluable visual aid. An important example of the use of
these straight-line approximations is in the design of linear control

systems.
roots. For complex conjugate roots, it is consymbols so that we can use the universal curves
adopt
standard
venient to

(d)

Complex conjugate

We describe the conjugate pole (zero) pair in terms

that result therefrom.

of a magnitude
as

shown

a>

measured from the negative

and an angle

in Fig. 8.19.

Explicitly, the

real axis,

parameters that describe the pole

which we call the undampedfrequency of oscillation,

known as the damping

(zero) positions are a>

and

= cos 0,
If

factor.

the

pole (zero) pair

is

given in terms of its real and imaginary

parts,
/>i,.

a and

ft

- - P

are related to

(8-20)

juo

following:

-J"o

g>

cos

to

P = <u

sin

o),

Vw*

and co by the

(8.21)

-/woVW 2

Returning to the definition of the

cos d, we see that

FIG. 8.19. Pole location in terms of

damping factor,

VI

*
and

<ot .

Network

226

the closer the angle

the angle 6

is

is

to

7r/2,

Bode

the

plots for the conjugate pole (zero) pair, let us set

The magnitude

for convenience.

20 log

|1

and the phase

is

a)

is the damping factor. When

damping factor is nearly unity.

the smaller

nearly zero degrees, the

To examine
>

and synthesis

analysis

+]2t,a>\

<f>(w)

then

is

= 20 log [(1 =

a> 8) 8

+ 4 <u 8]*
8

(8.22)

2gn>

-i

tan"

(8.23)

l-a>8

we examine the low- and high-frequency asymptotes of the magnitude,

we see that the low-frequency asymptote is decibels; the high-frequency
If

asymptote (for <o y> 1) is 40 log co, which is a straight line of 40

db/decade or 12 db/octave slope. The damping factor plays a significant
part in the closeness of the straight-line approximation, however. In
Fig. 8.20 the asymptotic approximation for a pair of conjugate poles
(a> = 1) is indicated by the dashed line. Curves showing the magnitude

We

1.0 are given by the solid lines.

0.6, and
0.1,
see
0.6 is the straight-line approximation a close one.
that only for

for

Universal curves for magnitude and phase are plotted in Figs. 8.21 and
8.22 for the frequency normalized function

G(*)

(sK)8

+ 2(s/a) ) +

(8.24)
1

We see that the phase response, as viewed from a semilog scale, is an odd
function about

eo/a>

The phase at co
1

<w

is

90 or n\2 radians.
1

+ 10

+5

^x^^-r-o.6y

-b

~~

Asymptote

-10
-15
0.1

0.5

1.0

Frequency,

^\

vft.

A.

0.2

^\

r-io-^N.

o>

2.0

5.0

FIG. 8.20. Magnitude versus frequency for second-order pole.

10.0

Amplitude, phase, and delay

qp'IfflOl-

227

228

Network

anal/sis

and synthesis

woaiv

8
CM

Amplitude, phase, and delay

229

For a conjugate pair of zeros, we need only reverse the signs on the scales
of the magnitude and phase curves.
KVraMpig 8.1. Using Bode plot asymptotes, let us construct the magnitude
versus frequency curve for
0.1s

GW

(8.25)

'(Ht

+
16X10*

10

two first-order break frequencies at o> - and a> 50. In

addition, there is a second-order break frequency at m 400. With a quick
calculation we find that 0.2 for the second-order factor. The asymptotes
are shown in Fig. 8.23. The magnitude and phase plots are given in Fig. 8.24
through a microfilm plot computer program.

We see

there are

-12db/oetave

FIG. 8.23. Asymptotes for G(s) in Eq. 8.25.

8.3

SINGLE-TUNED CIRCUITS

We will now study a class of circuits whose system functions can be

described by a pair of conjugate poles. These circuits are called singletuned circuits because they only need two reactive elements an inductor
and a capacitor. The undamped frequency of oscillation of the circuit is

then

= {LC)-

l/i
.

circuit in Fig. 8.25,

W=

H(S)

An

example of a single-tuned

whose

!o(i>

Vs)

circuit is the

R-L-C

voltage-ratio transfer function is

1/LC

1/sC

R+

sL

ljsC

s*

+ (K/L)s +

(8.26)

1/LC

230

Network

anal/sis

and synthesis

FIG. 8.24a. Magnitude of G(s) in Eq. 8.2S.

FIG. 8.246. Phase of G(s) in Eq. 8.2S.

FIG.

8.25. Single-tuned circuit.

Amplitude, phase, and delay

231

FIG. 8.26

The poles of H(s)

are

'

(8.27)

2\LC

2L

where we assume that (R*IL*)

J} J

< (4/LC).

In terms of a and

/J

in Eq. 8.27,

//(5) is

ff(s)

+^
(s + a+j/f)(s + a-7/8)
'

(8.28)

From the pole-zero diagram of H(s) shown in Fig. 8.26, we will determine
the amplitude response \H(Jm)\. Let us denote the vectors from the poles
to they'w axis as |MX and |M,| as seen in Fig. 8.26.
can then write

We

|HO)l
where

K = a* +

j8*

(8.29)

|MJ

|M,|

and
iMii
|M,|

- [ + ( + m*
- [a + (a, - W\"

In characterizing the amplitude response, the point

\H(jco)\ is

aspects.

(8.30)

eu

a>msx, at

which

maximum, is highly significant from both the analysis and design

Since \H(ja>)\

is

always positive, the point at which \H(Jco)\* is

point at which \H(ja>)\ is maximum.

maximum corresponds exactly to the

Since \H(jm)\* can be written as

Network

232

analysis

|H0>)|

and synthesis

+ F?
+ (a, + /?)*][* +(,-j8)8
(' +
8
4
> + 2tt>\%* - f) + (a +
('

[a*

(8.31)

/?*)'

/?*)

we can find a>max by taking the derivative of H(ja>)

setting the result equal to zero. Thus we have
|

dJH(U>)\*

__

<fo>

eoSuz

(8.32)

/**)

8
jS

(8.33)

(8.34)

Expressed in terms of the natural frequency of oscillation

damping factor

and

*
)l

f
dm

we determine

a>*

<8*)]

* |H

From the equation

with respect to

+ /8*)W + 2(q' + (* + /W
+ 2a "

(a'

*
\

and the

wj^ is

- (o) Vl

-*)*-

(C^o)

= <(\ ~ 2*)

(8-35)

Since comax must always be real, the condition for a>max to exist, i.e., the
condition for \H(jw)\ to possess a maximum, is given by the equation

21* <,

(8.36)

<

45.
cos 0, o)max does not exist for
so that t <, 0.707. Since
In
this
When 8 45, we have the limiting case for which max exists.
the
same
poles
have
0.707 and the real and imaginary parts of the
case,

magnitude,

i.e.,

/?,

or
>.

>

Vl

(8-37)

= 0. This is the lowest

> 0.707, or

We see from Eq. 8.34 that, when a /J, then o>max

frequency at which o)max may be located. For
>,

comax

is

imaginary;

it

> <Vl - P

<8 - 38 >

therefore does not exist To summarize, the key

that the imaginary part of the pole must be greater

point in this analysis is

or equal to the real part of the pole in order for a>max to exist. Interpreted
graphically, if we draw a circle in the s plane with the center at a and the
then the circle must intersect theyVo axis in order for comax
radius equal to
0,

as seen in Fig. 8.27. Moreover, the point at which the circle interfrom the triangle
sects the positive jm axis is comax- This is readily seen

to

exist,

Amplitude, phase, and delay

233

JU

max

>

<r

H-->
i

-"max

-70 x

FIG. 8.27. Peaking

with sides a,

/S,

By

a>max in Fig. 8.27.

circle.

the Pythagorean theorem,

we find

that
(oJLx

The circle described in

Fig. 8.27

is

- j8* - a*
called the peaking circle.

the peaking circle intersects the jot axis at

When a

>

/J,

(8.39)

o>

= 0,

the circle does not intersect the /to axis at

therefore, a>max cannot exist.

When

of the

a>

=w

/?,

all (Fig. 8.286);

the imaginary part of the pole

much greater than the real part, i.e., when /J J>

the/a> axis at approximately

When a

as seen in Fig. 8.28a.

is

then the circle intersects

the natural frequency of oscillation
a,

circuit (Fig. 8.28c).

A figure

of merit often used in describing the "peaking" of a tuned

Q, which is defined in pole-zero notation as

circuit is the circuit

(8.40)

2 cos

From this definition, we see that poles near the ja>

axis ( small) represent

high-g systems, as given in Fig. 8.28c, and poles far removed from the jo)
axis represent low-Q circuits (Fig. 8.28a). Although the Q of the circuit
given by the pole-zero plot of Fig. 8.286 is theoretically defined, it has no
practical significance because the circuit does not possess a maximum
point in

its

amplitude.

By means of
point,

which

\H(jw c)\

is

the peaking circle,

the frequency

- 0.707 |#(./aw)|.

co c

we can

also determine the half-power

at which the amplitude response

is

234

Network

and synthesis

analysis

ju

jco

JP

J0

Umax

j- *Jmm

a\

max

-jftk-

(a)

(6)

FIG. 8.28. Examples of peaking

undetermined,

(c)

circles,

a, eomax ~ p.

(a)

= P, ma

0.

(6)

>

fi,

We will now describe a method to obtain co c by geometrical construction.

Consider the triangle in Fig. 8.29, whose vertices are the poles {pi,pi*}
and a point to, on they'w axis. The area of the triangle is

= fa

Area (piPi*aid
In terms of the vectors
be expressed as

|MJ and

(8.41)

|M,| from the poles to

\MX

Area (&PiPi*<ot) =-

co it

the area can also

|M,| sin y

>

(8.42)

where

y> is

the angle at

it

as seen in the figure.

we
Ju

From Eqs.

8.41

see that the product |Mi|

|M 2

is

equal

to

|MX

|M,|

= -^
sin

\H(ja>i)\

where

(8.43)

y>

Since the amplitude response

is

(8.44)

K is a constant, then
.Ksin

FIG. 129

and 8.42,
|

|H(M)I

y>

(8.45)

20*

235

Amplitude, phase, and delay

For a given pole pair {puPi*} the parameters /J, a, and ^Tare prespecified.
we have derived \H(Jm)\ in terms of a single variable parameter,

Therefore,
the angle

y>.

When the angle y>

= tt\2 rad, then sin =

rp

|H(/>max)|

When v

w/4 rad, then sin

y>

\H(ja>J\

= coc

so that a>t
o) a
Let us
.

= Wmax, and,

(8.46)

= 0.707 and
= 0.707 I^C/o^nu)!

Let us consider

first

1, to,

now a geometric construction to

draw the peaking

circle as

shown

in Fig. 8.30.

obtain

We will

A the point at which the peaking circle intersects the positive

Now we draw a second circle with its center at A, and its radius
equal to AB, the distance from A to either one of the poles, as seen in
Fig. 8.30. The point where this second circle intersects the/a) axis is w c
denote by
real axis.

The reason is that,

at this point, the inscribed angle is

y>x

= w/4 because it

equal to one-half the intercepted arc, which, by construction, is w/2.

o)max
0, the half-power point m c is also called the half-power
bandwidth of the tuned circuit. In Fig. 8.31a the half-power point 's given
is

When

when

(Umax

= 0.

For a high-g

highly peaked at

\HQ0)\

m=

where comw
<*>< the amplitude is
shown in Fig. 8.316. In this case, if
are two half-power points m Ci and co
0i

circuit,

<Umx, as

< 0.707 |//(/ft>max)|, there

Peaking
'circle

FIG.

8.30.

Geometric construction to obtain half-power point.

Network

236

anal/sis

and synthesis

0.707

Iff,

(a)

FIG.

8.31. (a)

(b)

Low-G

circuit response,

(b)

High-Q

circuit response.

a>max, as seen in Fig. 8.316. By the construction process

we obtain the upper half-power point o> Ci It can be shown1
a>max is the geometric mean of m Ci and a> Ct that is,

about the point

just described,

that the point

(8.47)

As a

result, the

lower half-power point

>Oi

is

= a>mx

(8.48)

0>Ct

The bandwidth of the system for a high-g circuit of this type is described by

BW = m Ct

(o

(8.49)

0i

In design applications these high-g circuits are used as narrow bandpass

filters.

Finally, there are certain aspects

of the phase response of high-g circuits

that are readily apparent. In Fig. 8.32

we

see several steps in the process

is 0, as seen
of obtaining the phase response. The phase shift at eo
it rad, as shown in part (c)
oo, the phase shift is
from Fig. 8.32a. At <a
a> =i >max, the phase
of the figure. Finally, in the neighborhood of a
plane
is approximately
lower
half
in
the
shift resulting from the pole
is controlled
region
in
this
in
phase
change
The
(Fig.
8.32&).
tt/2
0,
in the
response
the
phase
seen
that
readily
It
is
the
pole
in large by
px
region of a> has the greatest negative slope, as seen from a typical phase

response of a high-g circuit shown in Fig. 8.33.

1

See for example, F. E. Terman, Electronic and Radio Engineering, McGraw-Hill

Book Company, New York,

1953.

Amplitude, phase, and delay

237

J<*

Hi

b)0

k
(*)

(b)

(e)

PIG. S.32. Several steps in obtaining phase response for high-Q

>

circuit,

() o><

a.

an example to illustrate our discussion of single-tuned

us find the amplitude response for the system function

Finally, as
circuits, let

#(s)

34

=
s*

Now we

determine the

and also the amplitudes

maximum and
|/f(/*comax)|

(8.50)

+ 6s + 34

and

half-power points

awx

and

<o c ,

\H(jo> c)\. In factored form, H(s) is

34

H{s)

(8.51)

+ 3-;5)
and the poles of H(s) are shown in Fig. 8.34. We next draw the peaking
circle with the center at j = 3 and the radius equal to 5. At the point
where the circle intersects thejfa> axis, we see that eomx 4. To check this
a* gives
result, the equation <*, =
cBmx = (5* - 3*)* = 4
(8.52)
(s

3+;5)(s

/J*

-180*-

FIG. 8.33. Phase response of high-Q

circuit.

Network

238

analysis

and synthesis

The amplitude
6.78

= <or
|H(/4)|

|f(/a>max)| is

then

34

(3+j9)(3-jl)

==

1.133

(8.53)

30

The

point

at

which the peaking circle

intersects the positive real axis is located

at s

= 2.0.

draw a

With the center

of radius

circle

AB

at A,

we

(equal to

At the point C where

new circle intersects theyeo axis, we
have o) c By measurement, we find
5-\/2 in this case).

this

to c

FIG. 8.34. Peaking

tion example.

~ 6.78

(8.54)

Let us check this result. Referring to

Fig. 8.34, we know_that the line segment

circle construc-

AB is of length 5>/2 it follows that AC

The line segment AO is of length
(8.55)
AO = 5 3 = 2 units
;

is

also 5-v/2 units long.

_.

Then

co c is

given as
co c

Finally,

we

sl(AC)1

|H(j'6.782)|
n

8.4

is

V46

6.782

(8.56)

(8.57)

obtain \H(j<o c)\ as

1

which

- (AO)* =

34
/

V(34

46)*

(6V46)

.
S

0.802

precisely 0.707 \H(j(Ow**l)\-

DOUBLE-TUNED CIRCUITS

In Section
poles.

8.3,

we studied the frequency response for a pair of conjugate

Now we turn our

attention to the amplitude response of

circuit.

pairs

We will consider

V2 (s)

FIG. 8.35. Double-tuned

two

we analyze here is the

of conjugate poles in a high-g situation. The

double-tuned or stagger-tuned circuit given in Fig. 8.35.
circuit

Amplitude, phase, and delay

239

the special case when the R, L, and C elements in the primary circuit are
equal in value to their counterparts in the secondary. Since the primary
and secondary inductances are equal, the mutual inductance is

M=KL

(8.58)

In this analysis we assume the coefficient of coupling

to be a variable
parameter. Let us determine the amplitude response for the voltage-ratio
transfer function V^V^s). From the mesh equations

V^s)

=^R +

L+

/.(s)

J^j

- sM /2(s)
(8.59)

= -sM h(s) +

{r

sL

+ -M

/,(*)

we readily determine

& + WQs + 1/LC]* -

Vito

Using tuned-circuit notation, we

K*

l
'

set

=-

2<o

(8.61)

H(s) can then be written as

H(s)

^RM'L

(1

K*)(s*
\

+ -^2_ , + --)( f + -fe_ s + _22jL\

1 + K
l + K/\
l-K 1-KI
(8.62)

If

we

*M

A=

set

I?(l

then

we can

^
K*)

= ^2*
1 - K*

(8.63)
K

'

write

H(s)

=
(s

where

{ Sl , Sl *}

11

SiXs

sfXs

s)(s

^S_ jJL+ K J "ll + K

(8.64)

s,*)

(1

-_]
K)*J
(8.65)

-k

Li

-x

ci

jc)*J

Let us restrict our analysis to a high- Q circuit so that *

1 . Furthermore,
let us assume that the circuit is loosely coupled so that AT 1. Under

Network

240

analysis

these assumptions,

and synthesis

we can approximate

the pole locations by discarding

Then the poles

the terms involving * under the radicals in Eq. 8.65.

approximately as
fai *i*} can be given

(8.66)
Similarly, {s *,*}

can be given as
{s s,*}

The

~ -o> jo) ^l +
J

pole-zero diagram of H(s)

poles

<u

is

(8.67)

given in Fig. 8.36. The real part of the

comparison to the imaginary parts for

is greatly enlarged in

FIG. 8.M. Poles and zeros of a dottble-tuned

circuit.

Amplitude, phase, and delay

Note that we have a triple zero at the

of the vectors in Fig. 8.36, the amplitude response is

clarity purposes.

origin.

^4|Mo|'

241

In terms

(8.68)

]H(J(0}1

JM.I |M,| |M.| IMtl

Since the circuit

is

high- 2 in the vicinity of

a>

= a>

we have

|M,|-|M4 |=i2|M |-2o,

so that in the neighborhood of

(8.69)

<u

Aco

|HO)|S'
4IMJIM.I
It is evident that

(8.70)

the amplitude response

of \H(jm)\ in the neighborhood of a>

depends only upon the pair of vectors

IMJ and |M,|. The double-tuned problem has thus been reduced to a singletuned problem in the neighborhood of
Consequently, we can use all the
on the peaking circle that were
derived in Section 8.3. Let us draw a
o)

results

peaking

circle

with the center at

- +/>,

(8.71)

FIG. 8.37. Peaking circle for doubletuned circuit.

and with a radius equal to tw^T/2, as

shown in Fig. 8.37. The inscribed

y> then determines the location of the maxima and half-power points
of the response. Without going into the derivation, the amplitude response
can be expressed as a function of y according to the equation

angle

|HO)l

A<o sin
4a)

sin

K(ta>

2(1

Referring to the peaking circle in Fig. 8.37,

let

(8.72)

K*)

us consider the following

situations:

> WiKfl:

In this case, the peaking circle never intersects the jot

axis; y> is always less than ir/2 (Fig. 8.38a), and the amplitude response
never attains the theoretical maximum
1.

a>

iW =

*-=

2(1

-K*)

as seen by the curve labeled (a) in Fig. 8.39. In this case,

circuit is said to

be undercoupled.

(8.73)

K < 2, and the

Network

242

anal/sis

and synthesis
jo>

8lM
2

\
i

-fo
T

7
WO

y
A

ly\ w

x^^
(a)

FIG. 8.38.

2. o>,

single point

(b)

(a)

Here the peaking

m=

(Fig. 8.38ft).

At

lmax

(c)

Undercoupling. (6) Critical coupling,

co^Kjl:

(c)

Overcoupling.

circle intersects

the amplitude

the/eo axis at a

equal to #max
in Eq. 8.73. In this case
Kj2, and we have critical coupling.
3. <u
cogA/2 The peaking circle intersects theytu axis at two points.
a>

a>

is

<

o> x

and

to,,

as seen in Fig. 8.38c.

The

intersecting points are given

by the

equation
<*>!,*

max

tt>

>

(8.74)

.[(!H*
(

Consequently, the amplitude response attains the theoretical maximum

Hum* at two points, as shown by curve (c) in Fig. 8.39. In this situation,
the circuit is said to be overcoupled.

Note that
have their

and critically coupled curves

The overcoupled curve, however, is

in Fig. 8.39 the undercoupled

maximum

points at

me

"a mix
FIG. 8.39.

(a)

Undercoupled

case, (A) Critically

coupled case,

(c)

Overcoupled case.

Amplitude, phase, and delay

uCl

uCt

<<

FIG. 8.40. Half-power points of overcoupled

maximum at o>! and u>t

we can determine
struction

critical

coupling,

the half-power points by using the geometrical con-

method given in Section 8.3. Observe that there are two

co c and co c
as shown in the overcoupled curve in
The bandwidth of the circuit is then

half-power points
Fig. 8.40.

circuit.

In the case of overcoupling and

243

BW =
The

Example &2.

a>

Ct

m Ci

voltage-ratio transfer function of

(8.75)

a double-tuned

circuit is

given as

As?

=
(*

+2

-HylOOX*

+2

+2

-ylOOX*

+yl06X>

+2

-/106)

(8/76)

From H(s), let us determine the following: (a) the maximum points <ot ma and
<o t ; (b) the 3 db bandwidth BW; (c) the damping factor ; (d) the coefficient
of coupling*; (e) the gain constant A; and (/) the maximum of the amplitude
response //max.
Solution, (a) The natural frequency of oscillation o> is taken to be approximately halfway between the two poles, that is, <o = 103 radians. In the neighborhood of , we draw the poles s 2 +/100 and s = 2 +/106, as
shown in Fig. 8.41 . From the peaking circle centered at the point s = 2 + /<.,

shown

in Fig. 8.41,

we

obtain

mt max
so that

(b)

mo =

<i

max

*o

m*r

Next we draw a

2* = 2.236 radians

+ 2.236 =
2.236 =

circle centered at

<o

V(3^2)

(8.77)

105.236 radians
(8.78)

100.764 radians

we have

this circle intersects theyVu axis,

">c,

"^3*

+ja> with radius 3V2. Where

Ct so that
1

- 4.123 radians

(8.79)

Network

244

and synthesis

anal/sis

FIG. 8.41. Peaking

The

db bandwidth

is

(c) The damping factor

from which we obtain

The

coefficient

which

oj)

{ is obtained

8.2.

103

of coupling

8.246 radians

from the real part of the poles

t
4

(d)

Example

then

BW - 20Ol -

circle,

circle for

u0.0194
'" 15^

(8.80)
a>

2,

(8.81)

K is obtained from the radius of the peaking

is

o>oA:

(8.82)

K = = 0.0582

We thus have
(e)

The gain constant

is

equal to

2{.A:

*
(f) Finally, the

- K*

maximum

(8.83)

2(2X0.0582)
1

- (0.0582)*

amplitude

flmiz

Hmmx

= ""**

(8.84)

is

=
2(1

-**)

-0.5009

(8.85)

Amplitude, phase, and delay

ON

83

POLES

AND ZEROS AND

245

TIME DELAY

time delay? How do we relate it to frequency response? We

attempt to answer these questions in this section. First consider the
transfer function of pure delay
(8.86)
<r tT
H(s)

What is

will

any excitation e(f) produces an

which is delayed by time T with respect
shown by the Laplace transform relationship,

For a system described by Eq.

identical response signal e(t

to the excitation. This

Bis)

is

8.86,

T),

- C[e(/ -T)] = er' T C[e(0]

(8-87)

Let us examine the amplitude and phase response of the pure delay.
From the equation
*-***
< 8 - 88)

HO) -

we

obtain the amplitude response

\H(Jw)\

and the phase response

We

see that the delay

response, that

<f>(a))

T is

(8.89)

coT

(8.90)

equal to minus the derivative of the phase

is,

T= -

d<f>(w)

(8.91)

dco

The magnitude, phase and delay characteristics of H(ja>) = erlaT are

given in Fig. 8.42a, b, and c.
If we define delay as in Eq. 8.91, we can readily deduce that for the
response to be nearly identical to the excitation, the system amplitude
response should be constant, and its phase response should be linear over
the frequency range of interest. If the phase is not linear, we have what is
known as delay distortion. To visualize delay distortion more clearly,
M<)

A()

*te)

(a)

(h)

FIG. a.42. Amplitude, phase, and delay of ideal delay function,

(ft)

Phase, (c) Delay.

(e)

(a)

Amplitude,

Network

246

anal/sis

and synthesis

we recall from Fourier analysis that any signal is made up of different

frequency components. An ideal transmission system should delay each
frequency component equally. If the frequency components are delayed
by different amounts, the reconstruction of the output signal from its
Fourier components would produce a signal of different shape as the input.

For pulse

applications, delay distortion is

an

essential design considera-

tion.

Let us next examine how we relate delay, or envelope delay (as it is

sometimes called) to the poles and zeros of a transfer function. For any
transfer function

n (-)
H(s)=*tf

(8.92)

IT(-Pi)

= a

with zeros at z

jcot and poles

at/>,

<r

jco i , the phase for

real frequencies is
lr

<K<)

? tan-1 W
=
=2
.

ot

*-i

Envelope delay

J-l

1 to
.

tan"

+ CO,
x
'

(8.93)
a,

h o' + (c (of + A o* + (

(894)
co,)*

see that the shapes of the delay versus frequency characteristic are

the same for

all

poles and zeros.

the poles, positive delay.

The zeros

However,

transfer functions with zeros alone.

Now

let

pole atp

contribute "negative" delay;

do not have
Ls is the only

linear physical systems

The inductor H(s)

exception. Its phase

is #a>)
ir/2; thus the delay is zero.
us consider the delay due to one singularity, for example, a
a +jcoQ The delay due to the one pole is

The following points are

1.

.IL

-J

is

dm

We

fflj
*

pertinent:

The maximum delay due

to this pole

is

Am = and occurs at
about

co

<w

to
.

co

The delay

versus frequency curve

(8.96)

is

symmetric

247

Amplitude, phase, and delay

FIG.

2.

8.43.

The frequency

Graphic construction to obtain delay bandwidth.

at

which the delay

o>x

3.

The

"effective delay

simply 2a

graphically

The

co

bandwidth"

is

is

half the

maximum, or

l/2or

is

(8.97)

ff

then

eo

<r

<m<

to

cr

or

The upper and lower half-bandwidth points can be obtained

by drawing a circle with center at a> = co, and radius a
.

intersections of the circle with the jco axis are the half-bandwidth

shown in Fig. 8.43.

The product of the maximum delay and delay bandwidth

points, as
4.
2.

Thus,

if

we wish

to obtain large delay

theyeo axis, the effective delay bandwidth

5.

The delay of an

is

always

by placing zeros or poles near

is

then very narrow.

all-pass function is twice the delay

due to the poles

alone.

The delay versus frequency curve

We

for the pole

is

shown

in Fig. 8.44.

see that the delay-bandwidth concept is only useful for a

rough

approximation, since the delay versus frequency characteristic only falls

as 1/to. To calculate the delay versus frequency characteristic for a

a number of poles and zeros, it is convenient to

obtain the delay curves for the individual singularities and then to add the
separate delays. It is not as desirable to obtain the total phase response

transfer function with

and then

differentiate numerically.

Finally,

when

it

should be pointed out that envelope delay only has meaning

the phase response goes through the origin. If it does not, there

a frequency-shift component in addition to the delay that

account for analytically.

is

is

hard to

Network

248

wo -

4<?o

">o

analysis

and synthesis

3o wo - 2<ro wo ^o

FIG.

o +

8.44. Plot

ffo

o +

2<ro

wo +

3<ro

wo + 4o

<*

of delay versus frequency.

Problems
8.1

and

Find the poles and zeros of the impedances of the following networks
on a scaled s plane.

plot

nnnr*

|f-

4h

io

Z(s>-

la)
o

M/*

1(-

20

iM

ZM-

m
20

4h

ZW-

PROB. 8.1

&2 The

circuit

shown

in the figure

is

a shunt peaking

video amplifiers,
(a)

Show that the admittance

Y(s) is of the

K(s

form

- s^s s>
(s

-*,)

Express slt sa , and ss in terms of R, L, and C.

circuit often

used in

Amplitude, phase, and delay

249

10"* mhos, find

-10 -ylO, and Y(jB)
-10 +yi0, st
(6) When st
the values of R, L, and C and determine the numerical value of *,.

=c

Y(,>

:r

PROS.

Find the amplitude and phase response for the following functions and

83
sketch.

() F(s)

(*) F(s)

+K

+K

s
(c) F(s)

Note
8.4

id) F(s)

and o> are positive

Given the function

that

7T v

quantities.

<**">

C() +71*.)

determine the amplitude and phase of G(jm) in terms of ^, B, C, D.

the amplitude function is even and the phase function is odd.

Show that

85 By means of the vector method, sketch the amplitude and phase response
for

s-l

+0.5

j
(a) F(s)

sis

(C) F(*)

10)

s
(b)

F(s)

's*
s
(e)

F(s)

'

(/)

+2s +2

Plot

F(5)
s*

+1
-1

^)F(j) - (,/2xV+ 9)
8.6

on semilog paper
F(*)

the

F(5)

-2s +

5*

F(,)

- 2j + 5

-(TTW+T)
=

s*
(j

+2s +5

2X*

Bode

plots of magnitude

100(1

+ 0.5*)

s(s

F(s)

<*>

50(1
(1

+2)

+ 0.025jX1 + O-IJ)
+ 0.05*X1 + 0.01s)

1)

and phase for

250
8.7

Network
Plot

analysis

on semilog paper

()

W
8.8

and synthesis

F(*)

the

Bode

plots of magnitude

10005
5 10~ 5*

and phase for

(1

+ O.OOZsXl +

(1

200(1 + 0.05s)
+ 0.02jX1 + 4 10-** + 10-V)

+ 10~V)

For the function

*W-3 + 2s + 5
s*

determine o> mu , \F(jco m *x)\, the half-power point

amplitude and phase response.

a>

and

\F(ja>

)\.

Sketch the

8.9 For the circuit shown, determine the current ratio IJIg and find: (a)
the point <>,, where its amplitude is maximum; (b) the half-power point <oc ;
1. Use geometric construction.
(c) the point to where |/i(a) M)//B(<o u)|

2.5 h

PROS.
8.10

A network function consists of two poles at p

ltt

jco t , as given in the figure.

Af\to)

8.9

is

maximum

at

o>

Show

m* =

/<*

r^*^' -9

'

= at

that the square of the amplitude response

|cos 20|.

PROB.

8.10

8.11 In connection with Prob. 8.2 plot the poles and zeros of the impedance
Find, approximately, the maximum point of the
l/l%s).
function Z(s)
amplitude response. In addition, find the bandwidth at the half-power points
and the circuit Q.

Amplitude, phase, and delay

251

8.12 The pole configuration for a system function H(s) is given in the figure.
From the plot, calculate:
(a)

(6)

The undamped frequency of oscillation

The bandwidth and Q.

a>

yioo

-5

<T

-y'100

PROB.

8.12

In connection with Prob. 8.10 determine the ratio Af\m aM)lM*(0).

8.14 Determine the amplitude and phase response for the admittance Y(s)
of the circuit shown. Is the peaking circle applicable here? What can you say
about the shape of the amplitude response curve in a high-g situation? Determine the bandwidth of the circuit and the circuit Q.
8.13

20

oWW

lh

^SW"

YM-

=rif

PROB.

8.14

8.15 For the overcoupled case of a double-tuned circuit, derive an expression

for the peak-to-valley ratio that is, M(o> m x)/M(a> ), where A#(-) denotes amplitude. Use the notation in Section 8.4. {Hint: see Prob. 8.13.)
8.16

For the voltage

ratio of a double-tuned circuit

fri v\

(s

+ 4 +y50X* + 4 -y50X* + 4

+j6Q)(.s

+4

-y60)

252

Network

analysis

and synthesis

and
circle to determine the maximum and half-power points
coupling K.
the circuit Q. Find the gain constant A and the coefficient of
and halfa x i mu
8.17 For the double-tuned circuit shown, determine the
of
power points and the circuit Q. Find the gain constant A and the coefficient

Use the peaking

coupling K.
10

"9

M=5xl<r 6 h
PROB.
8.18

Determine the delay at

a =

8.17

0, 1,

and 2 for

()

F(s)

(b)

F(s)

(c)

F(s)

(<*)

F(s)

(e)

s+2
= s-3
s

+3

3*

=
(s

s*

s + 1
+ Is + 5

IX*

+ 2)

(s+2Ks*+2s+ 2)

chapter

Network

analysis

9
II

NETWORK FUNCTIONS

9.1

network theory, the word port has a special meaning. A

regarded as a pair of terminals in which the current into
one terminal equals the current out of the other. For the one-port network
shown in Fig. 9.1, 1=1'. A one-port network is completely specified
when the voltage-current relationship at the terminals of the port is given.
For example, if V = 10 v and 1 = 2 amp, then we know that the input
or driving-point impedance of the one-port is
In

electric

port

may be

Zi

= = sa

(9.1)

Whether the one-port is actually a single 5-Q resistor, two 2.5-Q resistors
in series, or two 10-Q resistors in parallel, is of little importance because
the primary concern

is

the current-voltage relationship at the port.

Consider the example in which /

admittance of the one-port is

Yto =

2s

V=

then the input

^ = 2s + 3

which corresponds to a 2-f capacitor in

its

3 and

(9.2)

parallel with

a $-Q

resistor in

simplest case (Fig. 9.2).

=!=2f

One-port
network

V
I'

FIG.

PIG. 9.2

9.1

253

io

254

Network

analysis and synthesis

Two-port
network

Vi

v2
2'

r
FIG. 9.3

Two-port parameters

A general two-port network, shown in Fig. 9.3, has two pairs of voltagecurrent relationships. The variables are Vx , V2 , Ju lz Two of these are
dependent variables ; the other two are independent variables. The number
of possible combinations generated by four variables taken two at a time
describing a two-port
is six. Thus there are six possible sets of equations
.

network.

We will discuss the four most useful descriptions here.

The z parameters

A particular set of equations that describe

z-parameter equations

Vx =

a two-port network are the

z^ + zj

(9 3)

V2 =

21 l

wJ%

In these equations the variables Vx and Vt are dependent, and Iv I2 are

independent. The individual z parameters are denned by
V,

z ll

J,=0

7,=0

(9.4)

v,
21

'M
/!=0

7,-0

observed that all the z parameters have the dimensions of impedance.

Moreover, the individual parameters are specified only when the current in
one of the ports is zero. This corresponds to one of ports being open
circuited, from which the z parameters also derive the name open-circuit
It is

Vi

zb

FIG. 9.4

Vi

Network

anal/sis

255

II

parameters. Note that z u relates the current and voltage in the 1-1' port
only; whereas z M gives the current-voltage relationship for the 2-2' port.

Such parameters are called open-circuit driving-point impedances. On

and z2l relate the voltage in one port to
the current in the other. These are known as (open-circuit) transfer
the other hand, the parameters z liS

impedances.
As an example,
in Fig. 9.4.

let

T circuit

us find the open-circuit parameters for the

We obtain the z parameters by inspection

Zll

+z

= zb + z

7,=0

z ti
7,-0

(9.5)
*1*

=Z

=Z

7,-0

_^2
*l

7=0

Observe that z ia
z 21
When the open-circuit transfer impedances of a
two-port network are equal, the network is reciprocal. It will be shown
later that most passive time-invariant networks are reciprocal. 1
Most two-port networks, whether passive or active, can be characterized
by a set of open-circuited parameters. Usually, the network is sufficiently
complicated so that we cannot obtain the z parameters by inspection, as
.

did for the T circuit in Fig. 9.4. The question is now, "How do we
obtain the z parameters for any circuit in general?" The procedure is as

we

We write

a set of node equations with the voltages at the ports

and other node voltages within the two-port Va V4
Vk
as the dependent variables. The independent variables are the currents It
and I& which we will take to be current sources. We then proceed to
write a set of node equations.
follows.

Vx

and

Vt

h = nn V + n^Vi H
- n.1^1 +
+
1

- "*iFi +
1

One important exception

is

r*
+ nu
lk V
rk
+ nSk V
rt
+ ns
Sk V

+ "**n

the gyrator discussed later in this chapter.

(9.6)

Network

256

analysis

and synthesis

where n it represents the admittance between the

If the circuit is

n it

=n

ft

equations,

As a

AM

= Gu +

made up of R-L-C

result, the ijth

and/th nodes, that

is,

sCu

ith

(9.7)
sL,il

elements only, then

it is

clear that

cofactor of the determinant of the node

must be equal to

they'ith cofactor,

ti ,

that

This result leads directly to the reciprocity condition z21

is,

A w = AH

= z^,

as

we

shall see.

Returning to the
and Vt We obtain

set

of node equations in Eq.

9.6, let

us solve for

Vx

Vl

~ A 1+ A

(9.8)

In relating this
parameters,

it is

last set

of equations to the defining equations for the z

clear that

u=

ii

An
A

ztt

= A8a

(9.9)

Since for a passive network

A^ = A w

it

follows that za

network is then reciprocal.

As an example, let us find the z parameters of the Pi
First, the node equations are

= z12

the

circuit in Fig. 9.5.

h"{TA + Yc)V -Yc V

1

Ia

CD

=-Yc V

(9.10)

+ (YB + Yc)Vt

Vi

v2

YB

FIG.

9S

d>

Network

The determinant

for this set of equations

anal/sis

II

257

is

Ay = YA YB + YA YC + YB YC

(9.11)

In terms of Ay, the open-circuit parameters for the Pi circuit are

+
^~ YBAy

Yc
Zai

~Ay
(9.12)

*
Now

Zu

Ay

YA + YC

Ay

us perform a delta-wye transformation for the circuits in Figs.

9.4 and 9.5. In other words, let us find relationships between the immittances of the two circuits so that they both have the same z parameters.
let

We readily obtain
Zl

-z

- z + z - ^

*11

- za + z -

-s-

r,

-T

YB + Yc

z
Zo

We then find

(9.13)

Ay
Ay

-&
Ay
(9.14)

Zc

Ay

The y parameters
Suppose we were to write a
Fig. 9.3.

Then the voltages

and the currents Ix and

set

of mesh equations for the two port in

Vx and V2 would become independent sources,

would be just two of the dependent mesh

mesh equations

currents. Consider the general set of

- uA + iA +
V% mA + "hJ* H
= msi/i+

Vi

= m uJ +
l

"

+ i
r-

m^/j

+ W.A

(9.15)

+ m kkTk

sum of the impedances in the /th mesh and m ti is

common impedance between mesh i and meshy. We note here again

where
the

it

represents the

258
that for

Network

anal/sis

and synthesis

m mH

an R-L-C network,

it

for all

and / Thus

reciprocity

holds.

Solving the set of mesh equations for /x and 7a ,

we

obtain the following

equations.

The equations of 9.16

h~ A

(9.16)

define the short-circuit admittance parameters as

h = VuVi + yit v*
h = y i^i + ynVt

(9

17 )

= A/A for all i andy.

Let us find the y parameters for the bridged-T circuit given in Fig. 9.6.
The mesh equations for the circuit are
where ytj

'.

=A+

(-

we

A=
A

\s

(9.18)

obtain

2(2s

-A

+ ~h
s

In straightforward fashion

l)h

1)

ls%

+ 4s +

(9.19)

-A - - ls + 2s +
*

FIG. 9.6

Network
The

analysis

259

II

short-circuit parameters are then

+ 4s + 1
2(2s + 1)
8
2s + 25+1
yi = Vi* 2(2s + 1)
When ylx = yn or zu = za2 the network is symmetrical. 2
yu

sfa

2s

(9.20)

Returning to Eq. 9.17, which defines the y parameters,

parameters are expressed explicitly as

911

we

see that the

_h
Vy Fi-0
/,

3/m

v* Fi-0

(9.21)

h
K
h

Vn

Vm

Ft-0

v* Fi-0

y parameters are also called short-circuit admittance

apparent. In obtaining yu and y a , the 2-2' port must
be short circuited, and when we find yn and ylt , the 1-1' port must be
short circuited, as shown in Figs. 9.7a and 9.76.
As a second example, let us obtain the y parameters of the Pi circuit in

The reason
parameters

Fig. 9.5.

that the

is

To

now

obtain y a and y M ,

we

short circuit terminals 2-2'.

We then

have

Vu
yi

yn, *2i

(9.22)

= -Yc

,Jz

m yn Z3.

h>i

(b)

(a)

FIG. 9.7

*A
and

symmetrical network

is

easily recognized because

by interchanging the

2-2' port designations, the network remains unchanged.

1-1'

Network

260

analysis

and synthesis

We next short-circuit terminals

1-1' to obtain

= YB + Yc
^12 = -Yc

^22

,g

23)

The h parameters

set

of parameters that are extremely useful in describing transistor

h parameters given by the equations

circuits are the

Vi
I

The

= *uA + hi*V*
"tl-M. "22'

individual parameters are defined

by the

relationships

-h
Fj=0

h
"21

7l-0

1*

=~

h
"89

see that h lt

and h 91 are

Jl-0

and hx% and h^

The parameter hn can be interpreted

short-circuit type parameters,

are open-circuit type parameters.

as the input impedance at port

seen that h u

is

(9.25)

=h

F,-0

We

(9.24)

"I"

1 with port 2 short circuited.

merely the reciprocal of ylt

It is easily

hu

(9.26)

yu
The parameter h 22

is

an open-circuit admittance parameter and

to z M by
ha

is

related

(9.27)

Both the remaining h parameters are transfer functions; h 21 is a shortand h 12 is an open-circuit voltage ratio. Their
relationships to the z and y parameters is discussed later in this chapter.
For the Pi circuit in Fig. 9.5, the h parameters are

circuit current ratio,

fcll

=
Ya

h 12

h ai

+ YC

Yc
YA + YC
= Yc
YA + YC
YA YC
v
YA +YC

fc_

(9.28)

Network

analysis

II

261

h
+

NIC

FIG.

9.8.

Negative impedance converter with load impedance.

hlt . This is the reciprocity

Observe that for the Pi circuit, h^
condition for the h parameters and can be derived from their relationships
to either the z or y parameters.
Next

let

us consider the h parameters of an ideal device called the

negative impedance converter (NIC), which converts a positive load

impedance into a negative impedance at its input port.* Consider the

NIC with a load impedance ZL shown in Fig. 9.8. Its input impedance is
Z.

= -Z,

(9.29)

= *

(9.30)

which can be rewritten as

i

The following voltage-current

relationships hold for the

NIC.

Vx - kVt
(9.31)

A = kh
If

we

interpret Eq. 9.31 using

h parameters, we arrive at the following

conditions.

h vt

(9.32)

21

We see that since h u

j&

h tl

the

NIC is nonreciprocal.
NIC is

In matrix notation, the h matrix of the

ii

(9.33)

i
.k

The NIC
is

is

a convenient device in the modeling of active

circuits.

not, however, a device that exists only in the imagination.

For a

Chapter

NIC, see L. P. Huelsman, Circuits,

and Linear Vector Spaces, McGraw-Hill Company, New York, 1963,

lucid discussion of the properties of the

Matrices,
4.

It

Practical

Network

262

realizations of

and synthesis

anal/sis

NIC's have been achieved using

an article by Larky. 4

transistors.

Some of

these are described in

The ABCD parameters

Let us take as the dependent variables the voltage and current at the
port

1,

and define the following equation.

LA.

(9.34)

ara

This matrix equation defines the A, B, C,

D parameters, whose matrix is

known as the transmission matrix because it relates the voltage and current
at the input port to their corresponding quantities at the output. The
reason the current It carries a negative sign

is

that

most transmission

engineers like to regard their output current as coming out of the output

port instead of going into the port, as per standard usage.

In explicit form, the ABCD parameters can be expressed as

A=

B= - Yi
h

lt=0

Fa=0

(9.35)

C=

r=o

From

these relations

we

see that

represents

an open-circuit voltage

a short-circuit transfer impedance; C is an opencircuit transfer admittance; and D is a short-circuit current ratio. Note
that all four parameters are transfer functions so that the term transmission matrix is a very appropriate one. Let us describe the short-circuit
transfer functions B and D in terms of y parameters, and the open-circuit
transfer functions A and C in terms of z parameters. Using straightforward algebraic operations, we obtain
transfer function; J?

is

A=^

B = _J_

J/21

i-

D = _2u

z 21

Vtl

(9.36)

c=
For the

ABCD

parameters, the reciprocity condition

is

expressed by the

equation

YA
det

B~]

C D

= AD-BC=l

* A. I. Larky, "Negative-Impedance Converters,'

CT-4, No. 3 (September 1957), 124-131.

Trans.

(9.37)

IRE on

Circuit Theory,

Network

263

II

Let us find, as an example, the

parameter for the ideal trans-

ABCD

former in Fig.

analysis

whose denning

9.9,

Vi

equations are

Vl - nVt
/1

<*SD

= i(_ j8)

FIG.

If we express Eq. 9.38 in matrix form,

9.9. Ideal transformer.

we have

n
(9.39)

so that the transmission matrix of the ideal transformer

B"

is

~n
^s

C D

(9.40)

I
L

Note, incidentally, that the ideal transformer does not possess an impedance or admittance matrix because the self- and mutual inductances
are infinite. 5

For the

ideal transformer terminated in a load

impedance shown in

Fig. 9.10a, the following set of equations apply.

(9.41)

nZ,
Taking the

ratio

of

Vx to I we find the input impedance at port

l3

Zl

Thus we

see that

an

==
h

n*ZL

ideal transformer is

the load element were an inductor

L (Fig.

to be

(9.42)

an impedance transformer. If
1 we would see

9.106), at port

an equivalent inductor of value n 2L. Similarly, a capacitor C at the load

would appear as a capacitor of value C\n* at port 1 (Fig. 9.10c).

As a second example
*

For a

indicating the use of the transmission matrix in

detailed discussion concerning ideal transformers, see

Introduction ta

Modem

Network

Synthesis,

M.

E.

John Wiley and Sons,

Van Valkenburg,

New

York, 1960.

Network

264

anal/sis and synthesis

h
m:li

v2

()

c
n?i,

c=

FIG. 9.10. Ideal transformer as an impedance transformer.

network

ABCD

analysis, consider the

Fig. 9.5.

Ys

parameters of the Pi circuit in

+Yc
Yc

B=

Yc
YA YB + YB YC + YA YC
C=
Yc
Y
+Y
A
C

(9.43)

If

we check for

reciprocity

AD-BC = (YA +

from Eq.

YC)(YB

we

9.43,

+ Yc) - (YA YB + YB YC + YA YC

a
Y
In

-&*!
9.2

see that

(9.44)

RELATIONSHIPS BETWEEN TWO-PORT PARAMETERS

The relationships between two-port parameters are quite easily obtained

because of the simple algebraic nature of the two-port equations. For
example, we have seen that /t u = l/y u and h sa = l/z 2 . To derive h lt in
terms of open-circuit parameters, consider the z parameter equations

when port

1 is

open

circuited:

= 0.
*\ = z
V = z22/2

/x

i2-'a

(9.45)

Network
Therefore

we have

= -y1

h lt

analysis

II

= 2H

265
(9.46)

V.
Similarly, since h

defined as a short-circuit type parameter,

is

we can

derive h 21 in terms of y parameters as

h a ==

**

(9.47)

We can express all the A parameters as functions of the z parameters or y

An easy way to accomplish this task is by finding out

parameters alone.

what the

and y parameters themselves.

and y parameters are not simply
reciprocals of each other (as the novice might guess), since one set of
parameters is defined for open-circuit conditions and the other for shortCertainly,

relationships are between the z

by

their very nature, the z

circuit.

The

and y

relationships

can be obtained very

easily

by using matrix

notation. If we define the z matrix as

[Z]

(9.48)
L*si

and the y matrix as

[11

yu yn\

(9.49)
J/ti

In simplified notation

we can write

the two sets of equations as

=
[/] =

[V]

and
Replacing

[7]

in Eq. 9.50

Vt]

[Z][7]

(9.50)

[Y][V]

(9.51)

by [Y][V], we obtain
[V]

[Z][Y)IV]

(9.52)

so that the product [Z][Y] must yield the unit matrix [U]. The matrices
[Y] and [Z] must therefore be inverses of each other, that is,
1

[ZTr

From
z

the relationship,

and

[Y]

we can

[IT-^IZ]

(9.53)

find the relations between the individual

and y parameters.
Zjg

Zji

A*
A.

(9.54)

A.

Network

266

anal/sis and synthesis

TABLE

9.1

Matrix Conversion Table

[*]

z 12

z 22

*22

z ia

[A]

[71

s/

3/12

A12

AT

a,

A,

Am

Agg

3/a

3/u

Agi

>

A,

A*

Agg

Ay

3/n

J/12

Au

Au

/<

3/22

An
An

A*

2/21

An

J?

J?

3/12

B
An

AT

A12

Ah.

Agg

D
C
D

ft

w.

A,

A,

M
A,

ii

A,

z 12

A,

Z 22

*2

3/ii

3/u

2/21

A,

[A]

Z 22

3/ll

ii

A.

3/22

An

21

Z21

3/21

3/21

An

Agi

z22

3/u

AM

3/21

A21

Agi

*m

where A

Vii

3/21

= zuz g ZigZax
g

and

ii

= T~

Zia

J/22

=
(9.55)

where

Av =

^n2/ 22

Vi^/n-

2/12

Zg!=

Using these

?21

identities

we can

derive the h

or ABCD parameters in terms of either the z or y parameters. Table 9.1

provides a conversion table to facilitate the process. Note that in the

tMe
9.3

b. T

= AD- BC

(9.56)

TRANSFER FUNCTIONS USING TWO-PORT PARAMETERS

will examine how to determine driving-point and

of
a two-port by use of two-port parameters. These
transfer functions
broad categories. The first applies to two-ports
two
into
functions fall

In this section

we

Network

analysis

II

267

without load and source impedances.

These transfer functions can be

For example, let us derive
the expressions for the open-circuit voltage ratio VJiVx by using paramz
eters first and y parameters next. Consider the z parameter equations
for the two-port when port 2 is open circuited.
described by

means of z or y parameters

Vt =

alone.

z ai/j
(9.57)

Vx - z^
If we take the ratio of V, to

Vx we obtain
,

Yl

Vt

By

letting

%L

(9.58)

*ii

/2 of the second y parameter equation go to zero,

we

derive

the open-circuit voltage ratio as

(9.59)

V\

In similar manner
two-port as

we can

Via

derive the short-circuit current ratio of a

(9.60)

Is

and

_ _M

(9.61)

The open- and short-circuit transfer functions are not those we usually
deal with in practice, since there are frequently source and load impedances to account for. The second category of two-port transfer
functions are those including source or load impedances. These transfer
functions are functions of the two-port parameters z, h, or y and the
source and/or load impedance. For example, let us derive the transfer

IJVX of a two-port network that is terminated in a resistor

of/? ohms, as given in Fig. 9.1 1. For this two-port network, the following
equate apply.
admittance

*?

+
Vi

Two-port
network

FIG. 9.11

V2

268

Network

analysis

and synthesis

"~l

l-

Zll

^-^

+1*12*2

1+

s~~-

v2

Vi

l'o-

2'

FIG.

By

Two-port equivalent.

9.12.

eliminating the variable

Vt we obtain
,

yJR
V1 yu + 1/R
the same. YS1 is

rsl

and ytl are not

of the two-port network terminated in a

Note that

resistor

(9.63)

the transfer admittance

R, and y n

We must be

is

the transfer

make
admittance when port 2 is short circuited.
nature.
similar
this distinction in other cases of a
In order to solve for transfer functions of two-ports terminated at
either port by an impedance ZL , it is convenient to use the equivalent
circuit of the two-port network given in terms of its z parameters (Fig.
parameters (Fig. 9.13). The equivalent voltage sources z lg/a
9.12) or
careful to

y
and zIx in Fig. 9.12 are called controlled sources because they depend
6
upon a current or voltage somewhere in the network. Similarly, the
current sources y lt Vt and y^Yx are controlled sources. For the circuit
in Fig. 9.12, let us find the transfer impedance Zm = VJIU with port 2
"1

a.

1
1

y2iVil

Y2

y22

yi
'

1
I

FIG.

v2

9.13.

2'

Two-port equivalent.

For a lucid treatment of controlled

2nd Ed., McGraw-Hill Book Company,

sources, see E. J. Angelo, Electronic Circuits,

New York,

1964.

Network
terminated in a load impedance
the /* mesh,

ZL

If

write the

269

II

mesh equation

for

we have

-tA-(*ii
Since

we

anal/sis

+ Zx )li

(9.64)

Vt = ItZL we readily obtain

,

z -

r = ^rr

It also is clear that the current-ratio transfer

two-port network

< 9 - 65 >

function for the terminated

is

-*

h
In similar fashion,

we

"*
+ zz,

(9.66)

Z22

obtain the voltage-ratio transfer function for the

circuit represented in Fig. 9.13 as

Via.

(9.67)

^8 + yM
Next, suppose

we

are required to find the transfer function

the two-port network terminated at both ends, as

shown

VJV

for

in Fig. 9.14.

We first write the two mesh equations

(9.68)

Next,

we

solve for It to give

/t

+ 2ll)(^2 + m) - *1**21
Vt = RtI%, we may now arrive

(9.69)

(1

From

the equation

at the following

solution.

Y*=_Ml =
V,

?t
{R1

z^)(R t

+ z,,) - znZlt

*u

r^aro >*
FIG. 9.14

(9.70)

Network

270

anal/sis and synthesis

WW^O1-^h

MAA"

Z22-Z12

211-Z12

(Z21-12Ml

v2

Z12

Vl

FIG. 9.15. Two-port equivalent

Note that the equivalent

circuit

with one controlled-voltage source.

of the two-ports in Figs. 9.12 and

circuits

9.13 are not unique. Two other examples are given in Figs. 9.15 and 9.16.
Observe that the controlled sources are nonzero in these equivalent
circuits

only

the circuit

if

nonreciprocal.

is

Finally, let us consider the hybrid equivalent circuit

shown in

Fig. 9.17.

Observe the voltage-controlled source A gl F2 at port 1 and the currentcontrolled source Ag^ at port 2. Let us find the input impedance Z ln
.

The

pertinent equations are

Vi

+ hi*V*

h lxh

(9.71)

Va = -ZLh = ~(h 2lh + h^V^ ZL

and

Solving Eq. 9.72 for

(9.72)

Vt we find
,

V =

htiZiJi
1

Substituting V, in Eq. 9.73 into Eq. 9.71,

t^

(9.73)

h&ZL
we have

hiah z &

( i.
\

(9.74)

h^LjJ
(9.75)

so that

fc

MZ
h

yi2

o-ii-

Vi

-WW.yu + yn

-<

,73
y22

+ yi2<

Q)

Vi

FIG. 9.1*. Two-port equivalent

circuit

with one controlled-current source.

Network

FIG. 9.17. Hybrid equivalent

anal/sis

271

II

circuit.

We can easily check to see that Eq. 9.75 is dimensionally correct since Au
has the dimensions of impedance, h 22 is an admittance, and h 12 h tl are
,

dimensionless since they represent voltage and current ratios, respectively.

9.4

INTERCONNECTION OF TWO-PORTS

In this section

we

will consider various interconnections of two-ports.

We will see that when a pair of two-ports are cascaded, the overall transmission matrix is equal to the product of the individual transmission
matrices of the two-ports. When two two-ports are connected in series,
their z matrices add ; when they are connected in parallel, their y matrices
add.

First let us consider the case in

ports

Na and N

which we connect a pair of two-

in cascade or in tandem, as

shown

in Fig. 9.18.

We see

that

ra-ra

The transmission matrix equation

12a

+
Vi

for

tt

a is

ara

(9.77)

hb

+'

Vsa

Vn

(9.76)

*i

Nb

FIG. 9.18. Cascade connection of two ports.

v2
-

Network

272

analysis and synthesis

h r"

h'

9-*-

DC

Vi

L.

<

Za

V2

V2

'

*b

Vi'

Gyrator__
FIG. 9.19. Gyrator in tandem with

Correspondingly, for N

T circuit.

we have

(9.78)

Substituting the second matrix into the

first,

we

obtain

Ma-eara
We

see that the transmission matrix of the overall two-port

<network

is

simply the product of the transmission matrices of the individual twoports.

As an example,

let

us calculate the overall transmission matrix of a

T network shown in Fig. 9.19. The ideal

an impedance inversion device whose input impedance Zta is
its load impedance ZL by

gyrator 7 in tandem with a

gyrator

is

related to

Zm =

a*YL

21

(9.80)

Zi
The constant a

in Eq. 9.80

is

defined as the gyration resistance.

regard the gyrator as a two-port,

its

Vx =

If

we

denning equations are

a(-/)

h - -a vt

(9.81)

' B. D. H. Tellegen, "The Gyrator, a New Electric Network Element," Phillips

Research Repts., 3 (April 1948), 81-101, see also Huelsman, op. eit., pp. 140-148.

Network
so that the transmission matrix of the gyrator
"0

analysis

in

II

is

.a

We see that for the gyrator

AD-BC<=-1

(9.82)

a nonreciprocal device, although it is passive. 8

The overall transmission matrix of the configuration in Fig. 9.19 is
obtained by the product of the individual transmission matrices
Therefore the gyrator

is

za

C D

+z

V* + V. + V.

i
La
(9.83)

V* + Vc + V.

+ *

ah

az h

we check

If

the configuration in Fig. 9.19 for reciprocity,

we

see that for

transmission matrix in Eq. 9.83

AD-BC=-\
We

thus see that any reciprocal network connected in tandem with a

gyrator yields a configuration that

Next consider the

(9.84)

situation in

are connected in parallel, as

is

nonreciprocal.

which a pair of two-ports

shown

in Fig. 9.20.

Na

and

Let us find the

FIG. 9.20. Parallel connection of two ports.

*

Most gyrators are microwave

devices that depend

upon the Hall effect in

ferrites.

274

Network

anal/sis

and synthesis

lc

FIG. 9.21

y parameters for the overall two-port network. The matrix equations

for the individual two-ports are

V
V

L/J
and

From

UJ
Fig. 9.20,

we

J/llo

Vila'

-Vila

2/*2a-

~2/ll6

2/l2&

-2/216

2/226-

>1"

(9.85)

>"
(9.86)

UJ
must hold.

see that the following equations

~ V\a = V\h
h = Ao + Aft

V\

V*

^2o

'26
(9.87)

A = Ao + I

In connecting two-ports in series or in parallel,

we must be

careful

not altered when

connected in series or parallel with another two-port. For example, when
we connect the two-ports in Fig. 9.21 in parallel, the impedances Z, and
Zg will be short circuited. Therefore, to insure that a two-port network

that the individual character of a two-port network

is

does not interfere with the internal affairs of the other, ideal transformers
are used to provide the necessary isolation. In matrix notation, the sum
of the individual [/J matrices of the two-ports in parallel must equal the
[/] matrix of the overall two-port network. Thus we have

(9.88)

Vila

2/

2/21a

Vtia

J/116

2/126

2/226-

y**

Network
*

II

275

/&,

/la

anal/sis

JV

e>

V2a

Ideal

Vi

Vz

In

lib
C

Nb

Vu

Vu,

'-'

FIG. 9.22. Series connection of two ports.

so that the y parameters of the overall two-port network can be expressed

in terms of the y parameters of the individual two-ports as

If

we connect

|yii

y1

Ly

yJ

two-ports in

+
lytia +
[vna

series,

as

yita

Vtu

Vin
(9.89)
ytu>-

in Fig. 9.22, we can express

network in terms of the z param-

shown

the z parameters of the overall two-port

eters

+
y*ta +

Vub

of the individual two-ports as

pil

^t]

LZi

2tJ

+ 2m
z Mo + z*i J
Z

*11&
_ pUa + U
Zflb
Lzaia + Z

Zl*.

"l

(9.90)

222&-1

21ft

We may summarize by the following three points.

When

two-ports are connected in parallel, find the y parameters

and, from the y parameters, derive the other two-port parameters.
2. When two-ports are connected in series, it is usually easiest to find
the z parameters.
3. When two-ports are connected in tandem, the transmission matrix
1.

first,

generally easier to obtain.

is

As a

final

in Fig. 9.6.

example,

let

us find the y parameters of the bridged-T circuit

We see that the bridged-T circuit would be decomposed into

a parallel connection of two-ports, as shown in Fig. 9.23. Our task is to

The y parameters
find the y parameters of the two-ports
a and
b
of b are obtained by inspection and are

first

= y*u, = i
= y*n = s
Viiii
ym>

a is

T circuit so

that the z parameters can be obtained

(9.91)

by

inspection.

276

Network

analysis

and synthesis

These are
zllo

2Mo S +

(9.92)

Z]
ZlSn
*12o
*Mo

We then find the y parameters from the equations

Az

2s

(9.93)

yi2a = yio = t
2 12a

Since both JV and

now

we know that ylt yat

The y parameters for the bridged-T

obtained as
3/ii

tflt

+ Viu
= s(s + 1)
2s + 1

Vila

2s*

+ 4s + 1
+ 1)

2(2s

(9.94)
Vl2a

Vltb

sf_
2s

9.5

are symmetrical two-ports,

for the overall bridged-T circuit.

circuit are

2s

2s

2(2s

l)

2s*

INCIDENTAL DISSIPATION

As we have

seen, the system function H(s) of an

R-L-C network consists

of a ratio of polynomials whose coefficients are functions of the resistances,

inductances, and capacitances of the network. We have considered, up
to this point, that the inductors

and capacitors are

dissipationless;

i.e.,

Network

analysis

II

277

no parasitic resistances associated with the L and C. Since, at

high frequencies, parasitic losses do play an important role in governing
system performance, we must account for this incidental dissipation
there are

somehow.

An

effective way of accounting for parasitic resistances is to

the pure inductances and capacitances with incidental
dissipation by associating a resistance r, in series with every inductor L { ,

"load

down"

and, for every capacitor Ct , we associate a resistor whose admittance is t ,

g
as depicted in Fig. 9.24. Suppose we call the system function of the
network without parasitic dissipation (Fig. 9.24a) H(s), and the system
function of the "loaded" network H^s) (Fig. 9.246). Let us consider the
the relationship between H(s) and H^s) when the dissipation is uniform,
i.e.,

in a

manner such

that

Lt

(9.95)

Ct

where the constant a is real and positive.

When a network has uniform dissipation or is uniformly loaded, the
sum of impedances in any mesh of the unloaded network

if

<=

R +
4

sL{

(9.96)

becomes, after loading,

m'a

= JR, + sL +
t

= R + L&i +
{

r,

a)

(9.97)

C (s +
4

a)

Si

c,

L-vw-J

HM
Hi(s)
(a)

FIG. 9.24.

* H(s + a)
(b)

(a) Original

network.

(A)

Loaded network.

Network

278

analysis and synthesis

Jf
-It-

=*=**

2Q

1Q

<2h

FIG. 9.25

Similarly,

on node

basis, if the

original (unloaded)

network
n if

admittance between any two nodes of the

is

=G +

sCt

(9.98)

then the same node admittance after loading

'

=G +

sCt

=G +

Q(s

g(

is

(9.99)
t

+ a) +

+ a)

L,(s

Since any system function can be obtained through

tions,

H(s
H(s

+
+

it is

mesh or node equabecomes

readily seen that the original system function H(s)

) after the

network has been uniformly loaded, that

is

/^(s)

a).

Consider the following example. Let us

the unloaded network in Fig. 9.25.
1

Vu

By
s

find the

y parameters for

we have

7s

12

= 1 + + - - 1 + 7
4

2s

Via

first

inspection

(9.100)

4s

= i=-4

Then, for a loading constant a = J, the loaded network is shown in Fig.

9.26. In parallel with the capacitor Cx = J f, we have an admittance
gl

In parallel with the capacitor

gt

= *C1 = $mho

(9.101)

Ct =

is

f,

the associated admittance

*Ca = imho

(9.102)

Network

anal/sis

II

279

80

-vw
L-\H
if

J
20-

6Q<

< 1Q

=*f

10

1 2h 1
3

= J.

FIG. 9.26. Loaded network a

In series with the inductor

L=
r,

h,

we have a

olL

resistor

=1Q

(9.103)

Now we determine the y parameters for the loaded network to be

1

>

= i + Ks + i) +
- 1 + A(* + t)
y'

K*

i)

=i+

+1 4
+
D* + 2
=i+
4(s + |)
2a-

(a

'-- (H~*H)

(9.104)

We

see that the y parameters of the loaded network could have been
obtained from the y parameters of the unloaded network by the relation-

ship

t (s)

H(s

a).

We will make use of the uniform loading concept to prove an important

theorem concerning network
9.6

realizability in

Chapter

10.

ANALYSIS OF LADDER NETWORKS

In this section we will consider a simple method of obtaining the

network functions of a ladder network in a single operation.* This
method depends only upon relationships that exist between the branch
* F. F. Kuo and G. H. Leichner, "An Iterative Method for Determining Ladder
Network Functions," Proc. IRE, 47, No. 10 (Oct. 1959), 1782-1783.

280

Network

analysis

and synthesis

z,

>

'i

*2

*
I

^*4

*J*i

FIG. 9.27. Ladder network.

currents

and node voltages of the

in Fig. 9.27, where

all

all

ladder.

Consider the network shown

the series branches are given as impedances and

If the v denote node

denote branch currents, then the following relationships

the parallel branches are given as admittances.

voltages

and

apply.
vi+1

il+i

Zi+s +

vi+>

These equations form the basis of the method we discuss here.

To illustrate this method, consider the network in Fig. 9.28, for which
the following node voltage and branch current relationships apply.

- yJA v^s)
VJL) - '*(*) U') + v&) = [1 + Z^*) Yt (s)] v&)
h(s) - Y2 (s) VJLs) + Us)
= {Y&)[1 + ZJs) Us)] + Y,{s)} VAs)
V (s) = I (s)Z (s)+Va(s)
= {zjlyai + z,yj + rj + (i+ z,n)} v&)
/(*)

(9.106)

Upon examining the set of equations given in Eq. 9.106, we see that
each equation depends upon the two previous equations only. The first
equation, Vt
V% is, of course, unnecessary. But, as we shall see later,
it is helpful as a starting point. In writing these equations, we begin at

+
Vi

Zi

\-^H

z3

Y2

Y*

T
v2

j.
FIG. 9.28

Network

analysis

II

281

the 2-2' port of the ladder

and work towards the 1-1' port. Each succeeding equation takes into account one new immittance. We see,
further, that with the exception of the first two equations, each subsequent
is obtained by multiplying the equation just preceding it by the
immittance that is next down the line, and then adding, to this product
the equation twice preceding it. For example, we see that /,( is obtained
by multiplying the preceding equation V&) by the admittance Yt(s).
The next immittance is ZJis). We obtain KB(i) by multiplying the previous
equation IJs) by ZJs) to obtain /gZjj then we add to this product the
equation twice preceding it, V^s), to obtain Va
It Z,
Vt The process

equation

then easily mechanized according to the following rules (1) alternate

writing node voltage and branch current equations; (2) the next equation
is obtained by multiplying the present equation by the next immittance
is

(as

we work from one port

to the other),

and adding to

this

product the

results of the previous equation.

Using

obtain the input impedance Z{n(s) by

by the equation for I^s). We obtain the
function Vs)IVi(s) by dividing the first equation

of equations,

this set

we

dividing the equation for V^s)

voltage-ratio transfer

P*C0 by the

last

equation

as transfer immittances

if

we

let

VJs) term is

K2(j) =

We obtain other network functions such

manner. Note that

In taking ratios of these equacanceled. Therefore, our analysis can be simplified

every equation contains

tions, the

V-fe).

and current

ratios in similar

Vt(s) as a factor.

1.

equation of the set were a current variable fj(s) instead of the

voltage Vi(s), the subsequent equations would contain the current variable
Is) as a factor, which we could also normalize to 7/s)
1. An example
If the

first

in which the

equation

a current rather than a voltage equation

may be seen by determining the y parameters of a two-port network.
Before we embark upon some numerical examples, it is important to
note that we must represent the series branches as impedances and the
shunt branches as admittances. Suppose the series branch consisted of a
resistor

R=

first

1 2

of the branch

is

in parallel with a capacitor

z(s) =

J f Then the impedance

.

-^L
= _i+R
+
llsC

and must be considered as a

ladder.

C=

is

Similarly, if a shunt

an inductor L^

(9 . 107'
v
)

single entity in writing the equations for the

branch consists of a resistor R1

branch is

= 2Q and

h, then the admittance of the

y(s)

7T~
2 + s

( 9 - 108 >

282

Network

analysis and synthesis

in this discussion is that we must use the total impedance

or admittance of a branch in writing the equations for the ladder.

The key point

Example 9.1. Let us find the voltage ratio VJVlt the current ratio Ijlu the
input impedance Zx = VJIlt and the transfer impedance Zn = VJI-i for the
network in Fig. 9.29. First, we must represent the series branches as impedances
and the shunt branches as admittances, as shown in Fig. 9.30. The branch
current and node voltage equations for the network are

h*-

-/fflnp-*

3s

3h

Vi

2i

Vi

V2

2f=t=

*h

FIG. 9.30

FIG. 9.29

V^s)

Us)

= 2s

Va(s) =

V,

3s(2s)

-(6s*
s

1)

6s*

(9.109)

It(s)

VM

+-

14s

14g+

+ l=6s* + 57+-l

6s2

\)

The various network

+ 2s =

S*

functions are then obtained.

(a)

6^+57^+8
3
14s

s*

Yi

<*)

+2s

6s

57s8

8
(9.110)

2s*
(c)
li

** ~

id)

Example
network

9.2.

Find the

14s*

/x

+2

"

14s2

short-circuit admittance functions

y n and

j/2 i

for the

in Fig. 9.31. To obtain the short-circuit functions, we must short

circuit the 2-2' port. Then we represent the series branches as impedances and

the shunt branches as admittances so that the resulting network

is

given as

is

Network

rVW*-i

V.

analysis

II

283

I.

2a
Vi

2f

lh =j=2f

FIG. 9.31

The

illustrated in Fig. 9.32.

/,

Ja

pertinent equations are

=1
Va =2
= K2) + 1 = * +

(9.111)

2$*

l/5\

5/2

We now obtain our short-circuit functions as

2

12

7J + 5 +
h

yn

-,'

la

h
~^~~5

(9.112)

2
's.aa

Vi
2

+l

FIG. 9.32

number of other

on ladder networks have appeared

but a few, there are the works of
Bubnicki," Walker," and Dutta Roy. 14

contributions

in the recent literature.

Bashkow,10 O'Meara,11

To name

10

T. R. Bashkow, "A Note on Ladder Network Analysis," IRE Trans, on Circuit

Theory, CT-8, (June 1961), 168.
11
T. R. O'Meara, "Generating Arrays for Ladder Network Transfer Functions,"
IEEE Trans, on Circuit Theory, CT-10, (June, 1963), 285.
11
Z. Bubnicki, "Input Impedance and Transfer Function of a Ladder Network,"

IEEE Trans, on Circuit Theory, CT-10, (June,

"F. Walker, "The Topological Analysis

1963), 286.

of Non-Recurrent Ladder Networks,"

No. 7, (July, 1964), 860.

14
S. C. Dutta Roy, "Formulas for the Terminal Impedances and Transfer Functions
of General Multimesh Ladder Networks," Proc. IEEE, 52, No. 6, (June, 1964), 738.

Proc. IEEE, 52,


284

Network

analysis

>

and synthesis

Because of the recursive nature of the equations involved, the digital

ideal tool for the analysis of ladder networks. In Chapter
15 a detailed description is given of a digital computer program based
upon the algorithm described in this section for evaluating ladder network

computer is an

functions.

Problems
9.1

Find the

z,

y, h,

and

T parameters of the networks shown in the figures.

Some of the parameters may not be defined for particular

circuit configurations.

-OQ

PS
Y
'

PROB.
9.2

Find the

(e)

(b)

(a)

z,

y, h,

lo

and

9.1

T parameters for the networks shown in the figures,

iA/W

\AAA/

<
i

o2

Network

93 Find the z parameters for the lattice and bridge

shown. (The results should be identical.)

9.4

For the

lattice

and bridge

terms of the admittances

Ya =

analysis

and

Yb

MZb

285

circuits in the figures

circuits in Prob. 9.3 find the

1/Z

II

y parameters

in

93 For the circuit shown, find the voltage-ratio transfer function VJVX and
the input impedance x = V^lx in terms of the 2 parameters of the two-port
network
and the load resistor RL .

Z
1

I2

+
Vi

Rl<

v2
_

2'

1'

PROB. 9S

Network

286

anal/sis

and synthesis

9.6 For the cascade connection of two-ports depicted in the figure, show that
the transfer impedance Zj 8 of the overall circuit is given in terms of the a parameters of the individual two-ports by the equation

In addition, show that the short-circuit admittance yri

V-a

Vu.b

Vita

Na

Vi

Nb

PROB.
9.7
figure.

of

given by

VnJti

= -

is

Lx

v2
PROB.

9.6

9.7

Find the z and y parameters of the transformer (nonideal) shown in the

Determine the T- and w-equivalent circuits for the transformer in terms
L2 , and M. {Hint: Use the z parameters for the T-equivalent circuit,

and the y parameters for the w-equivalent circuit.)

9.8 The inverse hybrid parameters of a two-port network are defined by the
equation

eh

*ii

gl*
<f22j

g parameters in terms of either the z or y parameters and give a

physical interpretation of the meaning of these parameters; i.e., say whether a
parameter is an open- or short-circuit parameter, and whether it is a driving
point or transfer function. Finally, derive the conditions of reciprocity for the
g parameters.
Express the

9.9

Prove that for a passive reciprocal network

AD - BC =

where A, B, C, Dare the elements of the transmission matrix.

9.10 Find the z and h parameters of the
sented by its T-circuit model.

-VA

common

emitter transistor repre-

Wv-^O

-o

mh

J2

V2

Vi

PROB. 9.10

Network
9.11

anal/sis

II

287

The circuit in part (a) of the figure is to be described by an equivalent

shown in part (b). Determine Zeq in (b) as a function of the elements

input circuit

and voltages

in (a).

R,

9.12

of the

Find the

T parameters

of the configurations shown in parts (a) and (b)

figure.

9.13 Find the y parameters of the twin-T circuit in Prob. 9.2c by considering
the circuit to be made up of two T circuits in parallel.

9.14

Find the z parameters of the

circuits

shown.

Network

288
1

anal/sis

and synthesis

>

-o

V2

Vi

V
1

2*

Ideal

transformer

(a)

h<

V2

Vi

Ideal

<b)

transformer

2'

PROB.

9.14

PROB.

9.15

Ideal

transformer

PROB. 9.16

Network

analysis

II

289

9.15

Find the

9.16

Find the z parameters of the

9.17

For the circuit in Prob. 9.26 determine the y parameters of the uniformly

parameters of the circuit shown.

circuit

shown.

loaded circuit derived from the original circuit with the dissipation a =0.1.
Plot the poles and zeros of both cases.
9.18 Find the transfer impedance VJIX for the circuit in part (a) and the
voltage ratio VJV1 for the circuit in part (b). Plot the poles and zeros for the
transfer functions obtained.

o>- /Tnnn^-r-^voTP
}h

in

:4f

Vi

10

V2

(a)

WW-

Vi

2h

lf=:

if;*:

:ia

v2

PROB.

(6)

9.19

method

Find the

short-circuit parameters for the ladder

network

9.18

utilizing the

in Section 9.6.

10

HWv
20<

20

_lh

Hwvi-'Tnnp-^
=t=if

if

PROB.

9.1*

30

|VvV-|
Yf-rifippJfL-

20-

if

Vi

if==

1Q<

V2

4f=r

PROB. 9.20
9.20 Determine the voltage ratio VJVlt the current ratio IJIlt the transfer
impedance VJIlt and the driving point impedance Fi//a for the network shown.

chapter 10

Elements

<|>f

realizability

theory

10.1

CAUSALITY

AND

STABILITY

In the preceding chapters we have beep primarily concerned with the

problem of determining the response, giver^ the excitation and the network;
this problem lies in the domain of network analysis. In the next five
chapters we will be dealing with the problem of synthesizing a network
given the excitation E(s) and the response R(s).
any synthesis problem is the system function

The

starting point for

(10.1)

from a given system function.

procedur^
is to determine whether H(s)
step in a synthesis
There are two important
network.
physical
passive
can be realized as a
we mean that a
causality
stability.
By
and
causality
considerations

Our task is
The first

to synthesize a network

voltage cannot appear between any pair of terminals in the network

before a current is impressed, or vice ver^a. In other words, the impulse

response of the network must be zero for

h(t)

As an example,

is

causal, whereas

<

0, that is,

<0

(10.2)

e- (0

(10.3a)

the impulse response

h(t)

h(t)

= e- "
290

(10.36)

Elements of

realizability

theory

291

W-T)

(a)

FIG.

not causal.

is

(b)
(b) Realizable

Nonrealizable impulse response,

10.1. (a)

impulse response.

In certain cases, the impulse response could be made

by delaying it appropriately. For example, the impulse

realizable (causal)

response in Fig. 10.1a

seconds,

we

not realizable.

is

If

we

find that the delayed response h{t

delay the response by

T)

is

realizable (Fig.

10.16).

In the frequency domain, causality is implied when the Paley-Wiener

criterion 1 is satisfied for the amplitude function \H(jw)\. The Paley-Wiener
criterion states that a necessary and sufficient condition for an amplitude
function \H(ja>)\ to be realizable (causal)

is

that

dco

<

|log \H(ja>)\

+o>a

oo

(10.4)

The following conditions must be satisfied before the Paley-Wiener

criterion is valid: h(t) must possess a Fourier transform H(ja>); the
square magnitude function \H(jw)\* must be integrable, that is,
\H(jco)\ *dco

r.

<

oo

(10.5)

The physical implication of the Paley-Wiener criterion is that the amplitude

\H(jco)\ of a realizable network must not be zero over a finite band of
frequencies. Another way of looking at the Paley-Wiener criterion is that
the amplitude function cannot

fall off

to zero faster than exponential

For example, the ideal low-pass filter in Fig. 10.2 is not realizable
because beyond m c the amplitude is zero. The Gaussian shaped curve
order.

|HO)l
shown

in Fig. 10.3

is

(10.6)

not realizable because

|log|tf0a>)||

e~

o*

(10.7)

R. E. A. C. Paley and N. Wiener, "Fourier Transforms in the Complex Domain,"

Soc. Colloq. Pub., 19 (1934), 16-17.

Am. Math.

Network

292

analysis

and synthesis

\H(ja>)\

\H(ju)\

FIG.

FIG.

10.2. Ideal filter char-

f"
J-*

so that the integral

is

not

finite.

10.3.

Gaussian

filter

character-

istic.

acteristic

On the other hand,

>

(10.8)

dea

eof

the amplitude function

1

(10.9)

\H(jo>)\

vrr

co-

does represent a realizable network. Ih fact, the voltage-ratio transfer

function of the R-C network in Fig. 10.4 has an amplitude characteristic
given by \H(Jm)\ in Eq. 10.9.
For the ideal filter in Fig. 10.2, the inverse transform A(f) has the form

h{t)

=A

sin <o c t

(10.10)

nt

A is a constant. From the sin xfx curve in Fig. 3J4, we see that
nonzero for t less than zero. In fact, in order to make h(t) causal,
it must be delayed by an infinite amount. In practice, however, if we
delay h(i) by a large but finite amount t d such that for t < the magnitude
of h(t t^ is less than a very small quantity e, that is,
where
h(t) is

\Kt

Q\

<

*<0

wwv
1Q

Vi(s)

lr.^z

FIG,. 10.4

Elements of

we then can approximate

zero for

< 0.

h(t

(For a more

realizability

t by a causal response h^t) which

an

excellent treatment

then for a bounded excitation

If a network
words, if
other
In
bounded.
is
also
(0
is stable,

where
stable,

Cx and

C, are

then from

<

the convolution integral

<

the response

0<.t<co

C*

real, positive, finite quantities.

IKOI

by Wallman.*)

e(t)

0^r<oo

KOKCi
KOI

is

detailed discussion of the Paley-Wiener

criterion, the reader is referred to

then

293

theory

we

CiJ"|*(t)I dr

If a linear system

is

obtain

<

C,

(10.11)

Equation 10.1 1 requires that the impulse response be absolutely integrable,

or
""|/i(t)|

dr

<

(10.12)

oo

/;

One important requirement for h(t) to be absolutely integrable

impulse response approach zero as

approaches

is

that the

infinity, that is,

lim/i(r)-"0
<-00

can be said that with the exception of isolated impulses, the

impulse response must be bounded for all t, that is,
Generally,

it

|A(0I<C

all*

(10.13)

where C is a real, positive, finite number.

Observe that our definition of stability precludes such terms as sin ay
from the impulse response because sin a> f is not absolutely integrable.
These undamped sinusoidal terms are associated with simple poles on
the yw axis. Since pure L-C networks have system functions with simple
poles on the jm axis, and since we do not wish to call these networks
unstable, we say that a system is marginally stable if its impulse response
approach zero as t
is bounded according to Eq. 10.13, but does not
approaches infinity.
In the frequency domain, the

stability criterion requires that the

system

G. E. Valley Jr. and H. Wallman, Vacuum Tube Amplifiers, McGraw-Hill Book

Company, New York, 1948, Appendix A, pp. 721-727.

Network

294

anal/sis

and synthesis

function possess poles in the left-half piane or on theyw axis only. More-

on theyVo axis must be sini]pie. 3 As a result of the requirement of simple poles on they'w axis, if H{,s) is given as
over, the poles

H(s)

as

bm

m
s

+
+

a n_ 1 s

n~ 1

m~ l
b m _ lS

+
+! +

+a
bis + b

fl x s

(10.14)

then the order of the numerator n cannot exceed the order of the denominator m by more than unity, that is, n m < 1. If n exceeded m
by more than unity, this would imply that at s =jo> = oo, and there
would be a multiple pole. To summarise, in order for a network to be
stable, the following three conditions oii its system function H(s) must
be satisfied:
1.

H(s) cannot have poles in the right-fealf plane.

2.

H(s) cannot have multiple poles in

3.

The degree of

the denominator by
Finally,

it

tjhey'w axis.

the numerator of H(s) cannot exceed the degree of

more than

unity.

should be pointed out that a rational function H(s) with poles

in the left-half plane only has an inverse transform h(t), which

t

< 0.*

In

this respect, stability implies causality.

is

zero for

Since system funct'ons

of passive linear networks with lumped elements are rational functions

with poles in the left-half plane or jco axis only, causality ceases to be a
problem when we deal with system functions of this type. We are only

concerned with the problem of causality when we have to design a filter

for a given amplitude characteristic such as the ideal filter in Fig. 10.2.
We know we could never hope to realize exactly a filter of this type
because the impulse response would not^ be causal. To this extent the
Paley- Wiener criterion is helpful in denning the limits of our capability.

10.2

HURWITZ POLYNOMIALS

In Section 10.1
its

we saw

that in order fot a system function to be stable,

poles must be restricted to the left-half plane or theyeo axis. Moreover,

the poles

on

theyco axis

must be simple. The denominator polynomial of

the system function belongs to a class

*
i tt

In Chapter 6

it

of]

polynomials

was shown that multiple poles on

known

they'co axis

gave

as Hurwitz

rise to

terms as

sin wot.

4
G. Raisbeck, "A Definition of Passive Linear Networks in Terms of Time and
Energy," /. Appl. Phys., 25 (Dec., 1954), 15ip-1514. The proof follows straightforwardly from the properties of the Laplace transform.

Elements of
polynomials.

polynomial P(s)

is

theory

realizability

295

said to be Hurwitz if the following

conditions are satisfied:

1.

2.

is real when s is real.

The roots of P(s) have real

P(s)

As a

result

of these conditions,
P(s)

then

all

then

ac,

parts which are zero or negative.

must be

= (s+

Hurwitz because
hand,

is

all

real

if s {

<x

(10.15)

+ j@ is a root of P(s),

The polynomial

negative.

P(s)

a Hurwitz polynomial given by

= ansn + a^s"-1 +-- + a1s + a

the coefficients a {

must be

if P(s) is

1)(j

+js/2)(s

-js/2)

of its roots have negative real parts.

G(s)

(s

l)(.y

2)(j

(10.16)

On the

other

(10.17)

3)

not Hurwitz because of the root s = 1, which has a positive real part.
Hurwitz polynomials have the following properties:

is

1. All the coefficients a { are nonnegative.

This is readily seen by
examining the three types of roots that a Hurwitz polynomial might
have. These are

= y<
s = .jm
s cti.jPi
s

yt

a-i

The polynomial P(s) which contains

Since P(s)

is

= (s +

y< Xji

and

positive

and

positive

(Of real

P(s)

real

real

these roots can be written as

<*)[(*

oO*

the product of terms with only positive coefficients,

(10.18)
it

follows

must be positive. A corollary is that between

the highest order term in s and the lowest order term, none of the coeffithat the coefficients of P(s)
cients

may be

zero unless the polynomial

must not be zero

is

even of odd. In other words,

the polynomial

is neither even
nor odd. This is readily seen because the absence of a term a{ implies
cancellation brought about by a root s y< with a positive real part.
2. Both the odd and even parts of a Hurwitz polynomial P(s) have
roots on the ja> axis only. If we denote the odd part of P(s) as n(s) and
the even part as m(s), so that

fln-i. a-2>

fl <*i

P(s)

if

= n(s) + m(j)

(10.19)

Network

296

analysis and synthesis

then m(s) and n(s) both have roots on tne jw axis only. The reader is
8
referred to a proof of this property by Guillemin.
3. As a result of property 2, if P(s) is either even or odd, all its roots are

on

they'co axis.

The continued

4.

parts or the even to

quotient terms.
y(j)

= m(s)ln(s),

fraction expansion of the ratio of the

odd

odd

to even

parts of a Hurwitz polynomial yields all positive

n(s)lm(s) or
Suppose we denote the ratios as y>(s)
then the continued fraction expansion of y>(s) can be

written as
y(s)

= q xs +

(10.20)

+
ss

where the quotients qlt q%,..-,qn must be positive

6
P(s) = (j) + m(s) is Hurwitz.

sion,

we must perform a

series

To

if

the polynomial

obtain the continued fraction expan-

of long divisions. Suppose

V<s)

= 2

y>(s) is

(10.21)

n(s)
is of one higher degree than n(s). Then
obtain a single quotient and a remainder

where m(s)
m(s),

we

if

we

divide n(s) into

n(s)

The degree of the term R^s) is one lower than the degree of (*). Therefore
if we invert the remainder term and divide, we have
(L

Inverting

=q . + M2l

(io.23)

and dividing again, we obtain

RM^qj + M)
*,(*)

(10.24)

*.()

E. Guillemin, The Mathematics of Circuit Analysis, John Wiley and Sons, New York,
An excellent treatment of Hurwitz polynomials is given here.
*
proof can be undertaken in connection with L-C driving-point functions; see

1949.

E. Van Valkenburg, Introduction to Modern Network Synthesis, John Wiley and

Sons, New York, 1960.

M.

Elements of

We

realteability

theory

297

see that the process of obtaining the continued fraction expansion

simply involves division and inversion. At each step we obtain a

then invert the
quotient term q<s and a remainder term, R {+1 (s)/RJ[s).
into RJs) to obtain a new quotient.
remainder term and divide
There is a theorem in the theory of continued fraotions which states that

of

ip(s)

We

R^s)

the continued fraction expansion of the even to

odd or odd

to even parts

7
finite in length.

Another theorem states that, if

of a polynomial must be
odd
to even or even to odd parts
the
expansion
of
fraction
the continued
terms,
then the polynomial must
quotient
yields
positive
of a polynomial
JV(s).*
That is, if we write
factor
multiplicative
within
a
Hurwitz
to
be

fTO-irO^M

(10.25)

then F(s) is Hurwitz, if W(s) and F^s) are Hurwitz. For example,
test whether the polynomial
F(s)
is

- i* + s* +

5s*

3s

+4

let

us

(10.26)

Hurwitz. The even and odd parts of F(s) are

n(s)

We now

= s* +

3s

= "W/C0 by
and then inverting and dividing again, as given by

perform a continued fraction expansion of yfa)

dividing n(s)

by

m(s),

the operation
s*

+ 5s* + 4(s
s* + 3s*
2s* + 4)j + 3j(*/2
j + 2*

3s)s*

s)2s*

+ 4(2s

2s*

4)j(j/4

so that the continued fraction expansion of y(s)

*,)

= 2& - 5 +
()

is

(10.28)

s/4
'

See Van Valkenburg, be. eit.

W(s) is a common factor in mfc) and

n(*).

Network

298
Since

all

analysis

and synthesis

the quotient terms of the continued fraction expansion are

positive, F(s) is Hurwitz.

Example

10.1.

Let us

whether the polynomial

test

= 5s + 2s9 +

GO)
is

+6

3s

(10.29)

Hurwitz. The continued fraction expansion of n(s)lm(s)

is

obtained from the

division

2s2

6)5*
s*

+ 3* (s/2
+ 3s

We see that the division has been terminated abruptly

+ 3s. The polynomial can then be written as

by a

common

factor

5s

G(s)

We know
+ 3s is

s*

that the term

(s*

2/s

10.2.

is

+
>H)

3s)

which we referred to

= 57 + 2s* +

2s5

The term

(*)

wiO)

s*

3s

is

the mul-

earlier.

now

non-Hurwitz.

is

+ s* + 4s3 +

fraction expansion of F(s) is

(10.30)

Since the multiplicative factor

Hurwitz.

Next consider a case where W(s)

F(s)

The continued

Hurwitz.

is

also Hurwitz, then G(s)

tiplicative factor fV(s),

Example

85*

8*

+4

(10.31)

obtained.

1
'

fr

j^t)

(1 - 32)

(S4-+-4)

We thus see that

fV(s)

s*

W(s)
It is clear

that F(s)

Example 103.

is

+ 4, which can be factored into

= (s* + 2s + 2X*8 -2^+2)

(10.33)

not Hurwitz.

Let us consider a
F(s)

more obvious non-Hurwitz polynomial

= 5* + 5s +

2t*

35

+2

(10.34)

Elements of

The continued

realizability

theory

299

fraction expansion is
3

3s)S*
s*

+ 2s* + 2 (s
+ 3s*
-j*+2)j +

3*(-,j

5s)-5*+2(-,s/5
-4*
2}5iXfs

We see that F(j) is not Hurwitz because of the negative quotients.

Example

10.4. Consider the case where F(s) is an odd or even function. It is

impossible to perform a continued fraction expansion on the function as it
stands. However, we can test the ratio of F(s) to its derivative, F'(s)* If the
ratio F(s)IF'(s) gives a continued fraction expansion with all positive coefficients,

then F(s)

is

Hurwitz. For example,

F(s)

then F'(s)

F'(s)

is

if

F(s)

is

given as

= s7 + 34s + 2s3 + *
= 7s* + 15** + 6s +

(10.35)
1

(10.36)

Without going into the details, it can be shown that the continued fraction
expansion of F(s)IF'(s) does not yield all positive quotients. Therefore F(s) is
not Hurwitz.

POSITIVE REAL

10.3

In

this section

known

we

FUNCTIONS
of a class of functions
These functions are important because

will study the properties

as positive real functions.

they represent physically realizable passive driving-point immittances.

function F(s) is positive real (p.r.) if the following conditions are satisfied:

is, F(a) is real.

of F(s) is greater than or equal to zero when the real
greater than or equal to zero, that is,

1.

F(s)

is

2.

The

real part

part of s

is

real for real s; that

Re

[F(s)]

for

Re s

Let us consider a complex plane interpretation of a p.r. function.

Consider the s plane and the F(s) plane in Fig. 10.5. If F(s) is p.r., then a
point

map

<r

on

F(s) plane.
*

the positive real axis of the s plane

onto, a point

F(p^ which must be on the

would correspond

to,

or

positive real axis of the

In addition, a point st in the right half of the s plane would

See Guillemin, he.

cit.

Network

300

analysis

and synthesis

j\mF
JU
F(t) plane

plane

o
F(sO

(TO

F(ao)

FIG.

10.5.

Ref

Mapping of s plane onto F(s) plane.

map onto a point Ffo) in the right half of the F(s) plane.

In other words,
maps onto the
right half of the F(s) plane. The real axis of the s plane maps onto the
real axis of the F(s) plane.
A further restriction we will impose is that F(s) be rational. Consider
the following examples of p.r. functions:

for a positive real function, the right half of the s plane

= Ls (where L is

a real, positive number) is p.r. by definition.

function,
then L is an inductance.
impedance
If F(s)
an
2. F(s) = R (where R is real and positive) is p.r. by definition. If F(s)
is an impedance function, R is a resistance.
3. F(s) = K/s (K real and positive) is p.r. because, when s is real, F(s)
is real. In addition, when the real part of s is greater than zero, Re (s) =
1.

F(s)
is

a>0.
Re

Then
Therefore, F(s)

is p.r.

Ka

(f)"

If F(s) is

o*

>0

(10.37)

a>

an impedance

function, then the corre-

a capacitor of I IK farads.
We thus see that the basic passive impedances are
Similarly, it is clear that the admittances

sponding element

is

Y(s)

=K

7(5)

Y(s)

= K

p.r.

functions.

Ks
(10.38)

are positive real if

is

real

and

positive.

We now show that all driving-

point immittances of passive networks must be

p.r.

The proof depends

the following assertion: for a sinusoidal input, the average power

dissipated by a passive network is nonnegative. For the passive network

upon

Elements of
in Fig. 10.6, the average

power

Average power

by the network

dissipated

Re

realizability

[ZJtjco)]

|/|

theory

301

is

(10.39)

We then conclude that, for any passive network

Re

[Zjytt))]

We can now prove that for Re s =

(10.40)

^ 0, Re Z^o- + jm) ;> 0.

Consider

the network in Fig. 10.6, whose driving-

point impedance

is ZJ^s).
Let us load
the network with incidental dissipation

such that if the driving-point impedance

of the uniformly loaded network is

Passive

network

ZiW

ZxCO, then

Zfc)

= ZJis +

a)

where a, the dissipation constant,

impedance of a passive network,

is

real

Re Z^jco)
Re ZJa

so that

no.

(10.41)

and

positive.

10.6

Since Z^s)

is

the

(10.42)

+ jm) ^

(10.43)

Since a is an arbitrary real positive quantity, it can be taken to be a.

Thus the theorem is proved.
Next let us consider some useful properties of p.r. functions. The
proofs of these properties are given in Appendix D.
1.

If F(s) is p.r., then l/F(s)

driving-point impedance

admittance,

is

is

is

p.r.,

also p.r. This property implies that if a

then

its

reciprocal, the driving-point

also p.r.

functions is p.r. From an impedance standpoint,

two impedances are connected in series, the sum of the
impedances is p.r. An analogous situation holds for two admittances in
parallel. Note that the difference of two p.r. functions is not necessarily

The sum of p.r.

2.

we

see that if

p.r.

3.

for example, F(s)

The poles and

=s

1/s is

not

p.r.

zeros of a p.r. function cannot have positive real

i.e., they cannot be in the right half of the s plane.

Only simple poles with real positive residues can exist on the yco axis.
5. The poles and zeros of a p.r. function are real or occur in conjugate
pairs. We know that the poles and zeros of a network function are

parts,
4.

functions of the elements in the network. Since the elements themselves

are real, there cannot be complex poles or zeros without conjugates

because this would imply imaginary elements.

Network

302

The

6.

may

analysis and synthesis

highest powers of the numerator

differ at

zeros at 5

most by

and denominator polynomials

and

unity. This condition prohibits multiple poles

oo.

The lowest powers of the denominator and numerator polynomials

may differ by at most unity. This condition prevents the possibility of
7.

multiple poles or zeros at s

8.

The necessary and

real coefficients F(s) to

0.

sufficient conditions for

be

p.r.

a rational function with

are

(a) F(s)

must have no poles in the

(b) F(s)

may have only simple poles on theyco axis with real and positive

right-half plane.

residues.
(c)

Re F(jm)

for

Let us compare this

a.

all

new

definition with the original

one which requires

the two conditions.

1.

F(s)

2.

Re F(s)

is

real

when s is real.
0, when Re s ^

0.

In order to test condition 2 of the original definition, we must test

every single point in the right-half plane. In the alternate definition,

we test the behavior of F(s) along the

apparent that testing a function for the three conditions

condition (c) merely requires that

jm

axis.

It is

given by the alternate definition represents a considerable saving of

effort, except in simple cases as F(s)
1/j.

Let us examine the implications of each criterion of the second definition.

Condition

we test the denominator of F(s) for roots in

we must determine whether the denominator of

(a) requires that

the right-half plane,

i.e.,

is Hurwitz. This is readily accomplished through a continued fraction

expansion of the odd to even or even to odd parts of the denominator.
The second requirement condition (b) is tested by making a partial

F(s)

and checking whether the residues of the poles

on theya) axis are positive and real. Thus, if F(s) has a pair of poles at
s = ja> lt a partial fraction expansion gives terms of the form shown.

fraction expansion of F(s)

The

jCOi

+ jco^

residues of complex conjugate poles are themselves conjugates. If

as they must be in order for F(s) to be p.r. then

the residues are real

K = Ki* so that

JO)

+ JCOx

s*

(O*

Elements of

If

x is

found to be

reaiizability

positive, then F(s) "satisfies the

theory

303

second of the three

conditions.

In order to

test for the third

first find the real part of F(ja>)

let

we must
To do this,

condition for positive realness,

from the

original function F(s).

us consider a function F(s) given as a quotient of two polynomials

F(s)

(10.45)

6(s)

We

can separate the even parts from the odd parts of P(s) and Q(s) so

that F(s)

is

8( S )

JV 2(s)

where Mj(s) is an even function and Ns) is an odd function. F(s) is now
decomposed into its even and odd parts by multiplying both P(s) and
Q(s)

by A/g

- Ns so that
r(s)

_ M + NiM*- N*
Mt + JVa Ma - Nt
= M M8t - N * Nt
M 8 - Nt
i

We
x

MM

see that the products

MN

and

MN
s

are

odd

If

we

let s

= joy, we

Nx - Mj Nt
M* - Nt

(10.47)

JV,

Nt

are even functions, while

Therefore, the even part of F(s)

^-^

is

(10.48)

is

Odd

while the

and

functions.

Ev[F(s) ]
and the odd part of F(s)

[F(s)]

^VY'

see that the even part of

odd part of the polynomial

is

(10.49)

any polynomial

is real,

imaginary, so that if F(jw)

is

written as
F(ja>)
it is

clear that

and

Re

[F(jo>)]

Re

[F(ja>)]

j Im

[F(ja>)]

+ j Im [F(jo>)]

(10.50)

= Ev [F(s)] \_ja
= Odd [F(s)] \^, a

(10.51)

(10.52)

Therefore, to test for the third condition for positive realness,

we

determine the real part of F(ja>) by finding the even part of F(s) and then
joy. We then check to see whether Re F{joy)
for all <w.
letting s

Network

304

anal/sis and synthesis

A(u)

Single
root

FIG. 10.7

FIG. 10.8

The denominator of Re F(j(o)

a(ja>)*

That

is

always a positive quantity because

- JVaO)2 =

M^to)*

+ N ((o)* ^

(10.53)

an extra j or imaginary term in N^jco), which, when

1, so that the denominator of ReF(ja>) is the sum of
two squared numbers and is always positive. Therefore, our task resolves
into the problem of determining whether
is,

there

is

squared, gives

A(a>)

4 M O) MJjai) -

Ntito) NJJm)

(10.54)

If we call the preceding function A{m),

we see that A(a>) must not have

of the type shown in Fig. 10.7; i.e., A(co) must never
have single, real roots of to. However, A(a>) may have double roots
(Fig. 10.8), because A(t) need not become negative in this case.
As an example, consider the requirements for
positive, real roots

F(s)

s*

to be p.r.

First,

we know

the left-half plane or

that, in

on they'd)

+
bs

(10.55)

order for the poles and zeros to be in

axis, the coefficients a, b, c

must be greater

or equal to zero. Second, if *

0, then F(s) will possess poles on the
jco axis. We can then write F(s) as
F(s)

S*

+C

'

5*

(10.56)
C

We will show later that the coefficient a must also be zero when b = 0.
Let us proceed with the third requirement, namely,
the equation

Re F(ja>)

^ 0. From

- JVxO) NAjo>)
a(-(o* + c) + 1x0*^0
A(co) = (b a)afi + ac ;>

M^o)) Mfim)
we have
which
It is

of

simplifies to

(10.57)

(10.58a)
(10.586)

evident that in order to prevent A(m) from having positive real roots

eu,

b must be greater than or equal to

a, that

is,

b^a. As

result,

Elements of readability theory

when b

= 0,

fulfilled in
1. a, b,

2.

then a
0. To summarize, the conditions that must be
order for F(s) to be positive real are

^ 0.

c
a.

Fx(s) =

We see that

sr
is p.r.,

305

3s

\+

(10.59)

,
2

while the functions

F<S) =

(ia60)

TT1
+2
s

As a second example,

are not p.r.

us determine the conditions for the

let

biquadratic function

F(s)

TT^TT
+ b^ + b

<

ia62>

s"

We

will assume that the coefficients au Aq, b u b are all real,

to be p.r.
positive constants. Let us test whether F(s) is p.r. by testing each require-

ment of the second

definition.

of the denominator * x and b are positive, the

denominator must be Hurwitz. Second, if bx is positive, we have no poles
on the yeo axis. Therefore we can ignore the second condition.
The third condition can be checked by first finding the even part of
First, if the coefficients

F(s),

which

is

(s

_
The

s*

+
(* +

[(a

ft)

ti

bof

fliftjs'

ft x

a &.
(10.63)

real part of F(j(o) is then

Re

[F o1

0)
~ :y 18+
- W< - "
t
!

)" +
to
CO
O
+
(

flA

dO.64)

ti

We

see that the denominator of

Re

[F(jca)] is truly

always positive

so it remains for us to determine whether the numerator of

ever goes negative. Factoring the numerator, we obtain

m, t =

(a.

a 1b1

I^

[(flo

feo)

Re

_ aAf _ 4flA

[F(jco)]

(1Q65)

Network

306

analysis and synthesis

There are two situations in which

Re

[F(jco)]

does not have a simple real

root.

When the quantity under the radical sign of Eq.

10.65

is

zero (double,

(complex roots). In other words,

real root) or negative

+ *o) - AP - 4aA <.

+ b ) - aAP <, 4a<A>
(a + K) - cA ^
(a + b ) - A <, 2y/a^

(10.66)

Ko

or

( 10 - 67)

[(fl

If

then

*A ^ (V^ - yfb

or

(10.68)

(10.69)

)*

(10.70)

+ ^-aA<0
aA - (a + b ) <, 2VaA
< aA - (a +
fa, + * ) - aA <

If

(a,

then

but

aA ^ (\Za V*o)

so again

The second
when

2.

real root is

situation in
eo*

[F(jco)]

and

does not have a simple

when
(10.75)

(10.76)

we have

A - 0*o + *o) > 2VoA > (o + b - aA

aA > (V^ - VV)
)

Thus

We

(10.73)
(10.74)

+ *o) - aAl 8 - 4aA >

(a + b ) - dA <

[0*o

10.75

(10.72)

in Eq. 10.65 is negative so that the roots are imagi-

nary. This situation occurs

From Eq.

which Re

(10.71)

thus see that Eq. 10.70

is

a necessary and

(10.77)
(10.78)

sufficient condition for

biquadratic function to be positive real. If we have

aA = (V^o~-V6o>
we will have double zeros for
Consider the following example:

then

F(s) =

s
s

We see that

aA = 2
so that F(s)

is p.r.

(s/a

Re
g

+
T
+

00.79)

[F(jeo)]

2s

5s

+
T
+

~
25

(10-80)

16

- V^ )2 = (V25 - Vl6 )*

(10.81)

Elements of

realizability

theory

307

The examples
illustrate the

just given are, of course, special cases. But they do

procedure by which functions are tested for the p.r. property.

Let us consider a number of other helpful points by which a function

might be tested quickly. First, if F(s) has poles on theyw axis, a partial
fraction expansion will show if the residues of these poles are positive

and

real.

For example,
F(S)

+5
TTTT,
+ 1)
3s 8

<

ia82>

s(s*

has a pair of poles at s

j\. The partial fraction expansion of F(s),

-2s
5
+& = -rr7
+

= j is negative.

shows that the residue of the poles at s

is

not

(io.83)

Therefore F(s)

p.r.

and admittances of passive time-invariant networks

use of our knowledge of impedances
connected in series or parallel in our testing for the p.r. property. For
example, if Z^s) and Z^s) are passive impedances, then z\ connected in
parallel with Z, gives an overall impedance
Since impedances

are p.r. functions,

we can make

Z(s)

Since the connecting of the

the passivity of the network,
that if Fjis)

and FJs) are

i77^

= Zi(s) +

two impedances in

we know

that Z(s)

p.r. functions,

<

Z^s)

parallel has not affected

must

also be p.r.

p.r.

(1085)

Consequently, the functions

and
where a and

We see

then

m = JMM.
must also be

1084>

K are

real

and

F(s)

= -*s + a

(10.86)

F(s)

=
s + a

(10.87)

positive quantities,

must be

p.r.

We

then

observe that functions of the type

+a

a
fi

must be

p.r. also.

+a

(10.88)

Network

308

analysis

and synthesis

Finally, let us determine

whether

= _*L_
s + a

jr( s)

F(s)
V

two terms must

we conclude

(10.90)

+ a/Ks

Therefore, the

ctfKs are p.r.

be p.r. Since the reciprocal of a p.r. function

that F(s)

sum of
is

the

also p.r.,

is p.r.

ELEMENTARY SYNTHESIS PROCEDURES

The
is

s/K

s/K and

see that the terms

10.4

(10.89)

If we write F(s) as

is p.r.

we

o,K0

basic philosophy behind the synthesis of driving-point functions

up a p.r. function Z(j) into a sum of simpler p.r. functions

to break

Ziis), Z4s), .... ZJs), and then to synthesize these individual Z,(j) as
elements of the overall network whose driving-point impedance is Z(j).

Zis)

= Us) + Zs) +

---

+ Zn(s)

(10.91)

"breaking-up" process of the function Z(s) into the

One important restriction is that all Zj(s) must be
sum
p.r. Certainly, if all Z 4(s) were given to us, we could synthesize a network
whose driving-point impedance is Z(s) by simply connecting all the ZJs)
First, consider the

of functions Zs).

in series. However, if we were to start with Z(s) alone, how would we

decompose Z(s) to give us the individual Z/j)? Suppose Z() is given in

general as
""

+ ""-i* 1 + + ' + . Jfi>

1
+ hs + ba Q(s)
ms" + t^s"- +
= (that b = 0).
pole
at
s
Consider the case where Z(s) has a

W=

7( *

fl

-s "

fe

is,

divide P(s)

by

we can denote

Q(s) to give a quotient D/s

as

Z1(^)

and

Za(s).

Z(s)

= - + R(s)

(1(>

92)

and a remainder

R(s),

Let us

which

Z^s)

+ Z&)

previous discussions, we know that Z^

Are Z! and Z, p.r. ?
the p.r. criteria given previously.
Consider
p.r.
IsZa(5)p.r.?
is

From

(10.93)

D/s

ZgCO must have no poles in the right-half plane.

2. Poles of ZJLs) on the imaginary axis must be simple, and their
residues must be real and positive.
1.

3.

Re

[ZJJa>)]

for all m.

Elements of

realizabiiity

Let us examine these cases one by one. Criterion

1 is satisfied

the poles of ZJ&) are also poles of Us). Criterion 2

same argument.

theory

is satisfied

309
because

by

this

simple partial fraction expansion does not affect the

residues of the other poles. When s jo>, Re [Z(yw)
D/jco]
0.
Therefore we have

Re ZAjco)

From the
a- partial

foregoing discussion,

= Re Zjja>) ^

it is

fraction expansion can be

seen that

if

made such

the form Kfs and the other terms combined

(10.94)

Us) has a pole

at s

that one of the terms

still

remain

= 0,
is

of

p.r.

A similar argument shows that if Us) has a pole at s = oo (that

m = 1), we can divide the numerator by the denominator to give a
is,

quotient Ls and a remainder term R(s), again denoted as Z^s) and Zt(s).

Us)
Here ZJs)

is

= Ls + R(s) = Z^s) + Z^s).

also p.r. If

ZXjs)

has a pair of conjugate imaginary poles on

jto1 , then Z{s) can be

the imaginary axis, for example, poles at s

expanded into

(10.95)

partial fractions so that

(10.96)

Re

Here
so that

=Re (_i^L_\ =0

(_2*M

Zj(.y) is p.r.

Re [Z(/<u)] is minimum at some point m^ and if the value of

Re ZijcDf) = Kf as shown in Fig. 10.9, we can remove a constant K<, Kt
from Re [Z(ya>)] so that the remainder is still p.r. This is because Re [Z(/o>)]
Finally, if

still be greater than or equal to zero for all values of w.

Suppose we have a p.r. function Z(s), which is a driving-point impedance function. Let Z(s) be decomposed, as before, so that

will

Z(s)
ReZfjwi

74s)

ZJs)

(10.98)

Network

310

anal/sis

and synthesis
Zi(s)

ZzM

Z(s)
^

>

FIG. 10.10

where both 2^ and Zj are p.r. Now let us "remove" ZjCs) from Z(,s) to
give us a remainder Z^s). This removal process is illustrated in Fig.
10.10 and shows that removal corresponds to synthesis of ZjO).
Example

10.5.

Consider the following

Z(s)

function

p.r.

= s* +2s +6
s(s + 3)

(10.99)

We see that Z(s) has a pole at s = 0. A partial fraction expansion of Z(s) yields
*(*)

2
*

+3

-2&)+2fr)
If

(10.100)

we remove Zt(s) from Z(s), we obtain Z2(s), which can be shown by a resistor

in parallel with an inductor, as illustrated in Fig. 10.11.

*

1(

1Q
Z<s)

i"

Za

FIG.

10.11

Example 10.6
7s

+2

n>- 2j +4

(10.101)

where ]%?) is a p.r. function.

Let us synthesize the network by first removing min [Re
part of Y(jco) can be easily obtained as
8

Y(jco)].

The

real

14o>*

(10.102)

We

see that the

minimum of Re [Y(ja>)]

min [Re YQw)] =\. Let us then remove

occurs at

to

Yx =\ mho

= 0,

and

is

from

Y{s)

and denote

equal to

Elements of
the remainder as YJs), as
p.r.
is

because

shown

we have removed

theory

The remainder function Y^s) is

in Fig. 10.12.

only the

realizability

minimum

real part of YJjm).

Y^s)

obtained as
3s

(10.103)

It is readily

seen that YJp)

Thus the

capacitor.

final

is

made up of a J-ii resistor in series with a

is that shown in Fig. 10.13.

f-farad

network

:jo

^20

>20
Y*'}

"

Y(s)
>-

YM

"

=f

>

J
FIG. 10.12

Example

10.7.

FIG. 10.13

Consider the

p.r.

impedance
fo8

3**

Z{s)

3s

(10.104)

The real part of the function is a constant, equal to

of 1 il, we obtain (Fig. 10.14)

Z1(s)=Z(s)-l =
The

reciprocal of

+3s

6s*

unity.

3^ +
6s*

Removing a constant

+ 3s

(10.105)

Zt(s) is an admittance
no) -

which has a pole at s

expansion of Yjfs);

This pole

<x>.

fit

is

+3s

3s*

removed by finding the

r1(i)=25+

5?4n

WWW
10

Z(s)

(10.106)
1

Zi(s)

FIG. 10.14

partial fraction

(10107)

Network

312

analysis and synthesis

WWW
in

Z(3)

FIG. 10.15

and then by removing the term with the pole

2 farads in parallel with Ys) below (Fig.

Y&) The

reciprocal of YJs)

Tito

25

at *

10.15).

oo

YJs)

to give a capacitor of

is

now

obtained as

(10.108)

^--j-j

is

Zjfc)

which is, clearly, an inductor of 3 h in

network is shown in Fig. 10.16.

3*

(10.109)

series with

a capacitor of

farad.

The

final

www10

ZM

3h

Z\z!=.

FIG. 10.16

These examples are, of course, special cases of the driving-point synproblem. However, they do illustrate the basic techniques involved.
In the next chapter, we will discuss the problem of synthesizing a network
with two kinds of elements, either L-C, R-C, or R-L networks. The synthesis techniques involved, however, will be the same.

thesis

Problems
10.1

Test the following polynomials for the Hurwitz property.

()

s8

(ft)

5*

to

S1

+5* +2s + 2
+5* +5 + 1

+ 5s + 5* + 5
+ 45* + 55 + 2

to

5s

+25

V)

+5
+ 25* + 2/ + 5

Elements of readability theory

31

10.2 Determine whether the following functions are p.r. For the functions
with the denominator already factored, perform a partial fraction expansion
first.

(a)

F()

**

F(s)

(&)

IX*8

(*

+
+

2X
1X5

3*

(s

F(s)

(c)

F(s)

()

F(s)

+2s +4

2s*

(s

+4*

.*

5**

+ 4)
+ 3)

+4
+

3*

Suppose Fj(5> and Ft(s) are both p.r.

= FjKs) - FJ.3) is also p.r.

103
F(s)

+s
+

**

+ 2)

Discuss the conditions such that

10.4 Show that the product of two p.r. functions need not be p.r. Also show
that the ratio of one p.r. function to another may not be p.r. (Give one example

of each.)
10.5

GivenZfr)-

*+

*\

X for Z(s) to be a p.r. function?

(a)

What

(A)

Find A* for

(c)

Choose a numerical value for

10.6

are the restrictions

Re [ZQot)]

Prove that

if

on

to have a second-order zero at

10.7

Z()

Z(s) by
10.8

WW

positive real.

-j

first

= 0.

X and synthesize Z(s).

Zt(s) and Zt(s) are both p.r.,
ZUi -

must also be

co

is

p.r.

Determine min [Re Z(/o>)] and synthesize

removing min [ReZ(/co)].

Perform a continued fraction expansion on the ratio

W=

S8

+ 2j* + 3j + 1
+ 5s + 2* + 1

What does the continued expansion imply if Y(s) is the driving-point admittance
of a passive network ? Draw the network from the continued fraction.

3 4
1

Network

analysis

and synthesis

the
10.9 The following functions are impedance functions. Synthesize
impedances by successive removals of/ axis poles or by removing min [Re (/)].

()

+ 4s
7*T2
j + i

(*)

s(*

s8

2?
<c>

s*

2)

+4

27+1
+ 3s +
s + l

chapter

Synthesis of one-port networks

with two kinds of elements

In this chapter we will study methods for synthesizing one-port

networks with two kinds of elements. Since we have three elements to choose
from, the networks to be synthesized are either R-C, R-L, or L-C
networks.
We will proceed according to the following plan. First we will discuss the
properties of a particular type of one-port network,
synthesize it. Let us first examine some properties of
functions.

PROPERTIES OF

I.I

L-C

and then we

L-C

will

driving-point

IMMITTANCE FUNCTIONS

Consider the impedance Z(j) of a passive one-port network.

Let us

represent Z(s) as

Z(S)

where

N N
x,

M M
lt

are

odd

M^s)
i^

a (s)

+ JVj(s)
^i
+ N&)

are even parts of the numerator

parts.

The average power

/J J j-j
(U1)

and denominator, and

by the one-port is

dissipated

= J Re [Z(jd)] \I\*

Average power

(1 \ .2)

where J is the input current. For a pure reactive network, it is known

that
the power dissipated is zero. We therefore conclude that
the real part of
Z(/to)iszero; that

is

Re Z(
where

Ev Z(s)

Ev 2ija>)

= tf'('W)-^')^)
Mt\s)-Nt\s)
315

(11.3)
,,
n
(1L4)
,

Network

316

In order for

anal/sis

Ev Kjio)

and synthesis

= 0, that

MJlJm)

is,

either of the following cases

80'o>)

~ W) fUja>) =

must hold:

M
M

= - Nt
N
t =

(a)

(11.5)

(b)

In case

(a), Z(i) is

and in case

We

(11.6)

see

(A)

from

this

Z(s)

Z(s)

(H.7)

(H-8)

development the following two properties of L-C

functions:
'"
Zjj^s) or YxcC*) is tne ratio of even to odd or odd to even P013
nomials.
have only imaginary
2. Since both Mis) and Ns) are Hurwitz, they
or Y^s) are on
Z^s)
of
zeros
and
poles
the
that
follows
roots, and it

1.

the imaginary axis.

Consider the example of an

L-C immittance
a,*'

a*1

function given by

a,

(u

9)

know,
Let us examine the constraints on the coefficients a { and *,. We
coeffireal,
the
positive
to
be
first of all, that in order for the impedance
impedance
an
that
know
also
We
cients must be real and positive.
qo is
function cannot have multiple poles or zeros on theyw axis. Since
the
and
numerator
the
of
powers
highest
the
axis,
defined to be on theyw
if
example,
For
unity.
most,
at
differ
by,
can
polynomials
denominator
of
the
order
highest
the
then
2n,
is
numerator
the
of
order
highest
the

denominator can either be 2n - 1 so that there is a simple pole at s = oo,

s = oo.
or the order can be 2n + 1 so that there is a simple zero at
differ
by
can
Similarly, the lowest orders of numerator and denominator
Z(s)
at
of
zeros
or
at most unity, or else there would be multiple poles
s

= 0.

Another property of the numerator and denominator polynomials is

for example, the next
that if the highest power of the polynomial is In,
powers must
succeeding
the
and
In
2,
highest order term must be
any missing
be
cannot
There
through.
way
the
differ by two orders all

Synthesis of one-port networks

317

no two adjacent terms of either polynomial may differ

by more
than two powers. For example, for Z(s) given in Eq.
11.9, if ft, = 0,
then Z(s) will have poles when
terms,

i.e.,

V+
so that the poles will be at *

(^V

*-

*i*

and

(11.10)

at

"*"

= 0,1,2,3

dm)

It is clearly seen that

none of the poles sk are even on theyto axis, thus

one of the basic properties of an L-C immittance function.
From the properties given in Eq. 11.11, we can write a general L-C
impedance or admittance as
violating

Z(S)

Expanding Z(j) into

2(.)

+ 'Ofr' + *')(' + **)

+
tf coW + <*) .(, + ,,)...

Jft'

we

partial fractions,

< 1112>

obtain

= S + -2p-j + -*&- +

KkS

(1L13)

where the t are the residues of the poles. Since these poles are
all on
theyco axis, the residues must be real and positive in order for
Z* to be
positive real. Letting s =jm, we see that Z(;o>) has zero
real part, and
can thus be written as a pure reactance jX(o>). Thus we have

ft),

ft)

ft)

= J x(<)

(11.14)

Differentiating X(a>) with respect to

+K +

dco

Since

all

<o*

the residues

<

are positive,

dX(io)

a>,

(,,-cV
it is

IT*
A
is

similar development

also positive, that

is,

we have

seen that for an

(1U5)

L-C function,

shows that the derivative of Im [Y{jw)]

<

1116>

= B(m)

Network

318

and synthesis

anal/sis

Consider the following example. Z(s)

is

given as

+
_
^'(s'
+ co^ +
***

Letting s

= jco, we obtain

(11.18)

to*)

X{a>) as

= ; x(co) =

zo)

<o a*)

^^^ <}
Kco(-coa

+j

cog )

(11.19)

0,
Let us draw a curve of X(co) versus o>. Beginning with the zero at co
<o
as
encountered
let us examine the sequence of critical frequencies
next
the
positive,
always
increases. Since the slope of the X(io) curve is
large or
critical frequency we encounter is when X(co) becomes infinitely
from
goes
and
sign
changes
X(co)
a>
As we pass 2 ,
the pole is at a> 2

pass through any critical frequency,

to -. In general,
seen from the wayy X(to) is written in
as
of
sign,
there is always a change
through a> 2 , with the slope of X(a>)
pass
After
we
the last equation.
the next critical frequency is the zero
that
see
easy
to
is
positive,
it
always
Thus, if an impedance function is an L-C immittance, the poles

whenever we

at ,.

and zeros of the function must alternate. The particular X(a>) under
powers
discussion takes the form shown in Fig. 11.1. Since the highest
the
of the numerator and the denominator always differ by unity, and
oo,
s
=
and at
lowest powers also differ by one, we observe that at s =
frequency, whether a zero or a pole.
and a zero at
is a zero at s
external
called
oo
are
and s
s
are
frequencies
critical
critical frequencies, whereas the remaining finite
to t
and
a>
cog,
example,
previous
3,
referred to as internal. Thus, in the

there

is

always a

critical

For the example just discussed, there

= oo. The critical frequencies at s =

are internal critical frequencies.

admittance
Finally, let us summarize the properties of L-C impedance or
functions.
1.

ZiC(s)

or

YLC&s)

is

the ratio of

odd

nomials.

IM

FIG.

I.I

to even or even to

odd poly-

Synthesis of one-port networks

X(w)

+2

+1

+3

FIG.

I.Z

The poles and zeros are simple and lie on theyeo axis.
The poles and zeros interlace on they'eo axis.
The highest powers of numerator and denominator must

2.
3.

4.

unity; the lowest powers also differ

by

There must be either a zero or a pole at the origin and

5.

The following functions are not L-C

Ks(s

Z(5)

by

infinity.

for the reasons listed at the

left.

4)

+ l)(s + 3)
3
s + 4s + 5s
Z(s) =
3s* + 6s
2
K(s + l)(s* + 9)
z( =
2
(s + 2)(s + 10)
(s

differ

unity.

2.

1.

(11.20)

On

the other hand, the function Z(j) in Eq. 11.21,

diagram

is

shown

in Fig. 11.2,

is

_ 2(s +

1)(s

s(s*

11.2

SYNTHESIS OF

L-C

whose pole-zero

an L-C immittance.

9)
(11.21)

+ 4)

DRIVING-POINT IMMITTANCES

We saw in Section 11.1 that an L-C immittance is a positive real function

with poles and zeros on the jco axis only. The partial fraction expansion
of an

L-C function

is

expressed in general terms as

+ Kxs

(11.22)

cog

The synthesis is accomplished directly from the partial fraction expansion

by associating the individual terms in the expansion with network elements.
If F(s)

is

an impedance

l/Ko farads; the term

term I^/s represents a capacitor of

an inductance of K henrys, and the term

Z(s), then the

Kx s

is

Network

320

and synthesis

anal/sis

2X,

tf.h

Wi

o^HTHf

l0Q0>

'-vOOOv-'
2X1

'

2! h
7T5" n

Z(s).

FIG.

2^5/(5

o) 42) is

11.3

a parallel tank circuit that consists of a capacitor of

\f2Kf farads in parallel with an inductance of IKjcof. Thus a partial

fraction expansion of a general L-C impedance would yield the network
shown in Fig. 11.3. For example, consider the following L-C function.
2(s

_,.

Z(S)

+ l)(s* +
*- + 4)

9)

(11.23)

A partial fraction expansion of TUjs) gives

Z(s)

2s

+ * + -72-

We then obtain the synthesized network in Fig.

(11.24)

11.4.

method is based upon the elementary

synthesis procedure of removing poles on the jco axis. The advantage
with L-C functions is that all the poles of the function lie on the jco axis
so that we can remove all the poles simultaneously. Suppose F(s) in
Eq. 11.22 is an admittance Y(s). Then the partial fraction expansion of
Y(s) gives us a circuit consisting of parallel branches shown in Fig. 11.5.
For example,

The

partial fraction expansion

s( S
(s

+ 2)(s + 4)
+ l)(s* + 3)

HI
2h

if

Z<8)

FIG. 11.4

Synthesis of one-port networks

321

Klziz
Y(s),

FIG. 11.5

The

partial fraction expansion of Y(s) is

r(s)

is

is

=s+
s

s*

(11.26)
1

from which we synthesize the network shown in Fig. 11.6. The L-C
networks synthesized by partial fraction expansions are sometimes called
Foster-type networks. 1 The impedance form is sometimes called a Foster
series network and the admittance form is a Foster parallel network.
A useful property of L-C immittances is that the numerator and the
denominator always differ in degree by unity. Therefore, there is always
a zero or a pole at $ = oo. Suppose we consider the case of an L-C
impedance Z{s), whose numerator is of degree In and denominator
is of degree 2n I, giving Z(j) a pole at s = oo. We can remove this
pole by removing an impedance L^ so that the remainder function Z^s)
is still L-C:
ZjCO

The degree of the denominator of Z 2(j)

2/i

degree by

1.

degree

ZCO

is

LiS

2n

(11.27)
1,

but the numerator

is

of

because the numerator and denominator must differ in

oo. If we
Therefore, we see that Z 8 (s) has a zero at s

2,

have a pole at s = oo,

which we can again remove to give a capacitor C*s and a remainder
Ya(s), which is
invert

Zt(s)

to give

Yt(s) =

l/Z^s),

Ys(s)

will

Y&) = Y2(s) - C2s.

(11.28)

2h<

lf=t

YM.

if:

FIG.

a. M. Foster,

11.6

"A Reactance Theorem," Bell System Tech. J., No. 3 (1924), 259-267.


322

Network

analysis and synthesis

Ll

c*3=

Cizk

FIG.

FIG. 11.8

1.7

We readily see that F3(s) has a zero at s oo, which we can invert and
remove. This process continues until the remainder is zero. Each time
we remove a pole, we remove an inductor or a capacitor depending upon
whether the function is an impedance or an admittance. Note that the
final structure

of the network synthesized is a ladder whose series arms

whose shunt arms are capacitors, as shown in Fig. 1 1 .7.

are inductors and

Consider the following example.

We

see that Z(s) has a pole at s

+ 4s1 + 3
= oo, which

we can remove by

dividing the denominator into the numerator to give a quotient 2s

remainder

Z,(j), as

shown

Then we have

in Fig. 11.8.

Zt(s) -

Z(s)

s
= -44s +

2s

Observe that Z^s) has a zero at s

the pole at infinity.

y3(s),

as

may

first

and a

10s

+ 4s +
a

(11.30)
3

Inverting Z^s),

oo.

we

again remove

Then we realize a capacitor of J farad and a remainder

be seen in Fig. 11.9.

+
Y3(s)=Ys)-s= *f
s
4
4s + 10s
3

Removing
of | h and

the pole at s

oo of

Z^j)

llYs(s),

Zt(s) - Z3(s) -~s =

3

2h
o

Y3 (s)

if-

FIG.

1.9

gives a series inductor

FTT1
fs + 3

Ginr*

(11.31)

(11.32)

Synthesis of one-port networks

323

/toot^

2h
Remainder

Z4>
FIG. 11.10

in Fig. 11.10. The admittance y4(s) = 1/Z4(j) has a pole at

which we remove to give a capacitor of f farad and a remainder
y6(j) = 3/2y, which represents an inductor of f h. Removing this inductor gives us zero remainder. Our synthesis is therefore complete and
the final network is shown in Fig. 11.11.
Since we always remove a pole at s = oo by inverting the remainder
and dividing, we conclude that we can synthesize an L-C ladder network
by a continued fraction expansion. The quotients represent the poles at
s = oo, which we remove, and we invert the remainder successively until
the remainder is zero. For the previous example, the continued fraction

as

shown

oo,

expansion
s*

+ 4s2 +

is

3)25

2s

+
+

+
fo +
4s3 +

12s3

16s(2s<-->

6s
I0s)s*
s*

+
+

4s 2

3(ts <->

3)4s3

5.2
2->

is'

4s3

+
+

I0s(is<->Z
is
3(|$ <->

3.2
2s

3)2y(fj*->

2s

We see that the quotients of the continued fraction expansion give the
elements of the ladder network. Because the continued fraction expansion
-'innp

2h

fh

**4=

FIG.

f"

ffi
11.11

324

Network

anal/sis an

synthesis

always inverts each remainder and divides, the successive quotients alternate between Z and Y and then Z again, as shown in the preceding
expansion. If the

initial function is an impedance, the first quotient must

an impedance. When the first function is an admittance,
the first quotient is an admittance.
Since the lowest degrees of numerator and denominator of an L-C
admittance must differ by unity, it follows that there must be a zero or a
pole at s = 0. If we follow the same procedure we have just outlined,
and remove successively poles at s = 0, we will have an alternate realization in a ladder structure. To do this by continued fractions, we
arrange both numerator and denominator in ascending order and divide
the lowest power of the denominator into the lowest power of the numerator; then we invert the remainder and divide again. For example, in
the case of the impedance we have

necessarily be

(s*
i*

z( S)

The continued

+ l)(s* +
3)
j
z
;r
s(s* + 2)

(H.33)

fraction expansion to give the alternate realization

is

+ 58)3 + As* + s\3I2s <-> Z

3 + fs
fs* + s*)2s + ^(4/5$ *-> Y
2s + $s*
isP^s* + 5*(25/2s <-> Z

2s

|j

i
The final synthesized network is shown in Fig.

11.12. The ladder networks

Cauer ladder networks because W. Cauer2 discovered
the continued fraction method for synthesis of a passive network.
Note that for both the Foster and the
Cauer-form
realizations, the number of
o
|(
1(
L
L
elements is one greater than the number of
^f
f
internal critical frequencies, which we defined
5 h oj
oj

realized are called

5h

*p

*p

FIG.

1.12

previously as being

all

the poles

and zeros

of the function, excluding those at s and

00. Without going into the proof of the
s

Wilhelm Cauer, "The Realization of Impedances with Prescribed Frequency

Dependence," Arch. Electrotech., 15 (1926), 355-388.

Synthesis of one-port networks

325

can be said that both the Foster and the Cauer forms give
elements for a specified L-C driving-point
function. These realizations are sometimes known as canonical forms.
statement,

PROPERTIES OF R-C DRIVING-POINT IMPEDANCES

11.3

The

properties of

known

properties of

onto the

we

it

minimum number of

the

will

R-C driving-point impedances can be derived from

L-C functions by a process of mapping the ja> axis

a axis. 8 We

assume that

all

will

not resort to this formalism here. Instead,

driving-point functions that can be realized with

two kinds of elements can be realized in a Foster form. Based upon this
assumption we can derive all the pertinent properties of R-C or R-L
driving-point functions.

Let us consider

first

the properties of

R-C

driving-point impedance functions.

L-C impedance given in

obtain a Foster realization of an R-C impedance by
simply replacing all the inductances by resistances so that a general R-C
impedance could be represented as in Fig. 11.13. The R-C impedance,
as seen from Fig. 11.13, is
Referring to the series Foster form for an

Fig. 11.3,

we can

Z(s)

- ^ + K + -&_ + -&- +
s
s + a
s +
g
R a = K, C = \\K Rx = Kja
x

(11.34)

cr

where C
l/A^,
x
X,
lf and so on. In
order for Eq. 11.34 to represent an R-C driving-point impedance, the
constants K\ and at must be positive and real. From this development,
two major properties of R-C impedances are obtained, and are listed in
the following.
Ci

Hf-

IH
Co

-AMr-

R2

*i

FIG. 11.13

Sons,

E.

Van Valkenburg,

New York,

Introduction to

1960, pp. 140-145.

Modern Network

Synthesis,

John Wiley and

Network

326

and synthesis

anal/sis

R-C driving-point impedance are on the negative

can be shown from a parallel Foster form that the
poles of an R-C admittance function are also on the axis. We can thus
conclude that the zeros of an R-C impedance are also on the a axis.
2. The residues of the poles, Kit are real and positive. We shall see later
that this property does not apply to R-C admittances.
1.

The

poles of an

real (or) axis.

It

R-C impedances are on the a axis, let

along the a axis." To find the slope,

Since the poles and zeros of

us examine the slope of

dZ{a)lda,

we

first let

Z(c) with respect to

Z(<r)

a.

Z(<r)

a in Z(s), and then we take the

Thus we have

= + K. + -&- + _&_ +
a
a+
a + a
o-j

da =

and

It is clear

Let us

? H
*

a*

(a

now look

(11.35)

(a

atf

1-

o%?

^^^0
da

that

derivative of

(n
^^
36}

(11-37)

at the behavior of Z(s) at the

two points where the

a = <a and at
a = (o = oo. This is readily done by examining the general R-C network
in Fig. 11.13 at these two frequencies. At a = 0, (d-c), if the capacitor C
is in the circuit, it is an open circuit and there is a pole of Z(s) at a = 0.
real axis

If

is

and the imaginary

not in the

circuit,

axis intersect, namely, at

then Z(0)

is

simply the

sum of all

the resistances

in the circuit.

= R + R2 +

+ R
(11.38)
because all of the capacitors are open circuits at a = 0.
At a = oo, all the capacitors are short circuits. Thus, if R^ is in the
circuit, Z(oo) = R^. If R x is missing, then Z(oo) = 0. To summarize
Z(0)

these last

two statements, we have

f

Z(0)

oo,

present

=
C missing

Z(oo)

If

we examine

the

o,

Rn

missing

Rm

present

two cases for Z(0) and Z(oo), we

Z(0)

Z(oo)

see that

(11.39)

Synthesis of one-port networks

Next,

let

327

us see whether the poles and zeros of an R-C impedance

We have already established that the critical frequency

function alternate.
nearest the origin

must be a

zero.

must be a pole and the

Therefore, if Z(s)

Z( S)

(5!

(s

critical

frequency nearest a

oo

given as

is

+
+

ga)(s
ffOO

+
+

g4)
(11.40)
or,)

Then, if Z(y) is R-C, the singularity nearest the origin must be a pole
which we will assume to be at s = gx the singularity furthest from
the origin must be a zero, which we will take to be s = g
Let us plot
4
;

z , ff) =

(g
(g

versus

g,

Z(0)

equal to a positive constant

is

beginning at a

+
+

+
a^ia +
ga)(ff

g4)
(11.41)

a3)

and extending to a

Z(0)

= ^*

= oo.

At a

= 0,

(11.42)

Since the slope of Z(g) is always positive as

a increases, Z(g) must
increase until the pole s
is
reached
(Fig.
11.14). At a
x

= -a

Z(g) changes sign, and

reached.

= <V

We

= au

negative until the next critical frequency is

see that this next critical frequency must be the zero,
is

Since Z(g) increases for increasing

a, the third critical
frequency must be the pole s
g8 Because Z(g) changes sign at g
the final critical frequency must be the zero, s
g4 . Beyond a
g4 ,
the curve becomes asymptotic to Z(oo)
1.
From this analysis we see
that the poles and zeros of an R-C impedance must alternate so that for
*

Z()

a*

FIG. 11.14

Network

328

anal/sis

and synthesis

the case being considered

oo

In addition,

we see

>

<r4

>

>

<r8

or

>

Oi

(11.43)

0.

>1

that

(11.44)

a, ex.

which shows that Z(0) > Z(oo).

To summarize, the three properties we need to recognize an
impedance are:
1.

Poles and zeros

2.

The

lie

on the negative

The

An

residues of the poles

example of an

= oo must be a zero.

must be

R-C impedance
Z(s)

= (s +

real

and

positive.

is:

+ 4)(s +
+ 2)(s + 6)

l)(s

s(s

The following impedances

and they alternate.

must be a pole whereas

real axis,

singularity nearest to (or at) the origin

the singularity nearest to (or at) a

3.

R-C

8)

(11.45)

are not R-C.

= (s + l)(s + 8)
(s + 2)(s + 4)
(s + 2)(s + 4)
Z(s) =
(s + D
(s + l)(s + 2)
Z(s) =
s(s + 3)
Z(s)

(11.46)

Let us reexamine the partial fraction expansion of a general

impedance.
F(s)

= ^ + K + -^i- +
S + Ot
s
a

R-C

(11.47)

Instead of letting F(s) represent an impedance, consider the case where

F(s) is an admittance Y(s). If we associate the individual terms in the

expansion to network elements,

we

then obtain the network shown in

FIG. 11.15

Synthesis of one-port networks

329

We sec that an R-C impedance, ZBC(s), also can be realized

R-L admittance Y^is). All the properties of R-L admittances are

Fig. 11.15.

as an

the same as the properties of

to specify whether a function

R-L

R-C impedances.
is

to be realized as

It is therefore

important

an R-C impedance or an

admittance.

SYNTHESIS OF R-C IMPEDANCES

11.4

OR

R-L

ADMITTANCES

We

postulated in Section 11.3 that the Foster form realization exists

an R-C impedance or an R-L admittance. Since Foster networks are
synthesized by partial fraction expansions, the synthesis is accomplished
with ease. An important point to remember is that we must remote the
minimum real part of Z(/a>) in the partial fraction expansion. It can be
shown* that min [Re Z(/co)] = Z(oo), so that we have to remove Z(oo)
as a resistor in the partial fraction expansion. In cases where the numer-

for

ator is of lower degree than the denominator, Z(oo) = 0. When the

numerator and the denominator are of the same degree, then Z(oo) can
be obtained by dividing the denominator into the numerator. The
quotient is then Z(oo). Consider the following example.

F(s)

The

3(a

2)(s

s(s

partial fraction expansion of the

+ 4)

remainder function

(11.48)

3)

is

obtained as

(11.49)

where F(oo) = 3. If F(s) is an impedance Z(s), it must be an R-C impedance and it is realized in the series Foster form in Fig. 11.16. On the
other hand, if F(s) represents an admittance, we realize Y(s) as an R-L
network in the parallel Foster form (Fig. 11.17).

-\[

f
f

AV
30

10
If

Z(s)

FIG. 11.16
4

Van Valkenburg,

loc

cil.

330

Network

analysis

and synthesis

30.
*k:

Jo:
lh.

Y(s)

FIG. 11.17

An alternate method of synthesis is based on the following fact. If

we remove min Re [Z(yto)] = Z(oo) from Z(y), we create a zero at s = oo
for the remainder ZjO). If we invert Z x (j), we then have a pole at $ = oo,
which we can remove to give Za(s). Since min Re [ Y^jco)] = Yt(co), if
we remove Y2(co), we would have a zero at s = oo again, which we
again invert and remove. The process of extracting Z(oo) or r(oo) and
the removal of a pole of the reciprocal of the remainder involve dividing

we see that the whole

by a continued fraction expansion.

the numerator by the denominator. Consequently,

synthesis process can be resolved

The quotients represent the elements of a ladder network.

For example, the continued fraction expansion of F(s) in Eq.
j2

l&y

35*+

9s

3s)3s*

9s

24(3

24>*

11.48

is

+ 3s($s
s* + f*
is)9s

24(27

9s
24)is(s/72

4*

If F(s) is

an impedance Z(j), the resulting network is shown in Fig.

an admittance Y(s), we have the R-L network of Fig. 11.19.

11.18. If F(s) is

-Wv

30
Z(s)

270

jinpr^-

o:

If:

Y(s)_

FIG. 11.18

FIG. 11.19

Synthesis of one-port networks

33

J PROPERTIES OF R-L IMPEDANCES AND

R-C ADMITTANCES

II

The immittance
parallel Foster

that represents a series Foster

R-C admittance
F(s)

is

R-L impedance or a

given as

= KK s + K + -&- +
s + a

(11.50)

The

significant difference

pedance

R-C impedance and an R-L imR-C "tank"

between an

that the partial fraction expansion term for the

is

circuit is Kj(s + at); whereas, for the R-L impedance, the corresponding
term must be multiplied by an s in order to give an R-L tank circuit
consisting of a resistor in parallel with an inductor.
The properties of R-L impedance or R-C admittance functions can be

much

functions.

the same manner as the properties of R-C impedance

Without going into the derivation of the properties, the more

significant

ones are given in the following:

derived in

Poles and zeros of an

R-L impedance or R-C admittance are located

and they alternate.
2. The singularity nearest to (or at) the origin is a zero. The singularity
nearest to (or at) s = oo must be a pole.
3. The residues of the poles must be real and negative.
1

on

the negative real axis,

Because of the third property, a partial fraction expansion of an R-L

impedance function would yield terms as

Ki
s

(11.51)

This does not present any trouble, however, because the term above does
not represent an R-L impedance at all. To obtain the Foster form of an
R-L impedance, we will resort to the following artifice. Let us first expand
into partial fractions. If Z(s) is an R-L impedance, we will state
without proof here that the partial fraction expansion of 7^s)js yields
positive residues. 6 Thus, we have
2i_s)js

Z>=*2 + K +
a>

he.

Actually,

cit.

^-+-s

(11,52)

a{

ZBL{s)js has the properties of an R-C impedance;

see

Van Valkenburg

Network

332

and synthesis

analysis

-VW

ifi
40.

Z(s)

&={=
to:

FIG.

FIG. 11.20

where *o,
Z(s) in

Ku

K ^

the desired form

multiply both sides by s, we obtain

Consider the following function:

we

If

0.

11.21

for synthesis.

m-

2(s
(s

+ l)(s + 3)
+ 2)(s + 6)

(11.53)

an R-L impedance or an R-C admittance because it

of
the first two criteria cited. The partial fraction expansion

F(s) represents
satisfies

F{s)

is

F(s)

so

we see

F(s)js,

that the residues are negative.

on the other hand,

F(s)
s

The

(11.54)

partial fraction expansion of

is

_ 2(5 +
s(s +

If we multiply both sides

14

- 3)
2)(s + 6)

by

s,

we

4
t

+
2

(11.55)

obtain
is

i.

IX*

Is

+
s

(11.56)

an impedance Z(s), it is synthesized in series Foster

Y(s),
form, giving the R-L network in Fig. 11.20. If F(s) is an admittance
11.21.
Fig.
shown
in
then the resulting network is the R-C network
To synthesize an R-L impedance in ladder form, we make use of the
If we remove Z(0) from Z(s), the
fact that min Re [Z(Jm)] = Z(0).
If F(s) represents

remainder function

Zx(s) will have

a zero at s

FIG. 11.22

= 0.

After inverting Z^s),

Synthesis of one-port networks

333

YM
FIG. 11.23

we can then remove

the pole at s

0.

Since the value Z(0)

by dividing the lowest power of the denominator

is

obtained

power

into the lowest

term of the numerator, the synthesis could be carried out by a continued

fraction expansion by arranging the numerator and denominator polynomials in ascending order and then dividing. For example, the following
function is either an R-C impedance or an R-C admittance.

= 2(5 + l)(s + 3) = 6 + 8s + 2s*

12 + 8s + s*
(s + 2)(s + 6)
The continued
12

8*

fraction expansion of F(s)

is

+ s*)6 + is + 2s*(l
6 + 4s + js*
4s + fs )12 + Ss + s\3ls
12 + J
ls + s*)4s +
4s +
2

f**(f

iV)& +

**(49/5s

aS

OAAA

S*

As*

Y(s),

an impedance function, the resulting network is the R-L network

If, on the other hand, F(s) is an R-C admittance
the network is synthesized as in Fig. 11.23.

11.6

SYNTHESIS OF CERTAIN

If F(s) is

shown

in Fig. 11.22.

Under

certain conditions,

R-L-C

R-L-C

FUNCTIONS

driving-point functions

may be

syn-

thesized with the use of either partial fractions or continued fractions.

For example, the function

z(')~

4- 2s

VL+ it
+
S'

( 1L58)

334
is

Network

analysis

and synthesis

neither L-C, R-C, nor R-L. Nevertheless, the function can be synthesized

by continued
s*

fractions as shown.

+ s+

+ 2s + 2(l-Z
+ s+1
s + iy + s +
S2 + s

l>*
s*

i(s

+ l(s + \*-Z
s+l

1>

The network derived from

this expansion is given in Fig. 11.24.

In another case, the poles and zeros of the following admittance are
on the negative real axis, but they do not alternate.

Y(s)

The

An

Y(s)

is

(11.59)

partial fraction expansion for Y(s)

FIG. 11.24

Since one of the residues

= (s + 2)(s + 3)
(s + l)(s + 4)

negative,

=1+

s+1

we cannot

+4

all

is

(11.60)

use this expansion for

Y(s)ls and then

method would be to expand

multiply the whole expansion by s.
synthesis.

When we

alternate

multiply by

y(a)

s,

we

i
6

(11.61)

+4

obtain,

ito-*
2
+
V

(11.62)

+4

Y(s) also has a negative term. If we divide the denominator of

this negative term into the numerator, we can rid ourselves of any terms

Note that

with negative signs.

y(j)

3_(l__U + J!
2

\3

1/

h
6

+4

(11.63)

Synthesis of one-port networks

335

60-

lofa:

Y(s)

in

At 4=

FIG. 11.25

The network

that

realized

is

from the expanded function

is

given in Fig.

11.25.

expand Y(s) by continued fractions, we see that negative

However, we can expand Us) = l/l%s) by continued
fractions, although the expansion is not as simple or straightforward as
in the case of an R-C function, because we sometimes have to reverse the
If

we

try to

quotients result.

order of division to

make

fraction expansion of Z(s)

5s

the quotients

all positive.

The continued

is

+ s*)4 + 5s + s%i
4 + * + !**
|s + is*)6 +
6 +

5s

+ s%n/5s

is

Hi
i**

+ Hs
Asy^s

+ **(

*
s*)&s(6I15s

As we

see,

the division process giving the quotient of 1/3 involves a

reversal of the order of the polynomials involved.

network

is

given in Fig. 11.26.

WV

AAAr

-K-

:&

Z(t)

FIG. 11.26

The

resulting ladder

Network

336

and synthesis

anal/sis

In the beginning of this section,

conditions can an

R-L-C

it

was

stated that only

under special

driving-point function be synthesized with the

use of a ladder form or the Foster forms. These conditions are not
given here because they are rather involved. Instead, when a positive

and it is found that the function is not synthesizable

by using two kinds of elements only, it is suggested that a continued
fraction expansion or a partial fraction expansion be tried first.
real function is given,

Problems
11.1

(a)

Which of the following functions are L-C driving point impedances?

Why?
4X* +
A_,, " jQ ++ 9X*
+ 25)
8

(6) Synthesize the realizable

16)

(**

w=

There

is

113

no need to

fr*

8)

*fc+4)

L-C impedance.

+ 9Xi* + 25)
+ 4X* + 16)

1X5*

jCi"

calculate the element values of the four networks.

Synthesize the

L-C driving-point impedance

Z(s)

in the

1X*

two Foster and the two Cauer networks

that could be used to synthesize the following

7(

impedances in a Foster and a Cauer form.

Indicate the general /orm of the

11.2

(j

Z*S)

'

form shown in the figure,

and farads.

i.e.,

6s*

+ 425* + 48
18j + 48s

J+

determine the element values of the network

in henrys

=FC8

PROB.

II

11.4 There exists an L-C network with the same driving-point impedance as
the network shown in the figure. This alternate network should contain only
two elements. Find this network.

Synthesis of one-port networks

-nnnp
2h

337

lh

4=if

2W
If:

PROB.

11.5

The input impedance

11.4

for the network

2*

Zta=

shown

+2

"i+2s*+25 +

is an L-C network: (a) Find the expression for

Foster series form.

If

is

(6) Synthesize

in

2b

PROB.

II

&

11.6 Indicate which of the following functions are either R-C, R-L, or
impedance functions.
s*

+2s

(a)

z(,)

" FT4?~+1

(b)

ZW

(c)

Z(5)=

Z(4)

(e)

Z(5)

s*

+ 4* + 3

+ 45 + 3

?T6F+8
5*

+ 5s + 6

+,
j

+ Ss* + 6
** +3

L-C

338

Network

analysis

and synthesis

An impedance function has the pole-zero pattern shown in the figure.

Z( 2) = 3, synthesize the impedance in a Foster form and a Cauer form.
11.7

If

ju

-5

-3

-1

PROB.

11.7

11.8 From the following functions, pick out the ones which are
tances and synthesize in one Foster and one Cauer form.

+ l)Qr + 3)
+ 2X* + 4)
sjs + 4X* + 8)
is + lXs + 6)

= 4js +

2js

Y(s)

(s

sis

= (s +

Y(s)

l)fr

3)

+2)

l)fr

sis

R-C admit-

+ 4)

+2)

11.9 Find the networks for the following functions. Both Foster and ladder
forms are required.

Z(s)

().

is

(b)

3is

Z(s)

For the network shown,

find
1

2+Y
2s

Synthesize

+ 4)

sjs* +3)
+s* +6s +

Y as an L-C admittance.

PROB.
11.11

Y when

K9
=

4)

+2)
1)(*

s
11.10

1)(*

s(s

11.10

Synthesize by continued fractions the function

s

*w- 5

s*

+ 2s + 3j + 1
+ s* +2s + 1

Synthesis of one-port networks

339

11.12 Find the networks for the following functions in one Foster and one
Cauer form.

(s

Z(s)
11.13

2X*

20 +

+ 4)
+ 4)

0.5)(*

Synthesize the following functions in Cauer form.

+s + 1
+ J8 +s
s* + s*+2s + l
Z(s)
V+4 + 3*+a + l
4^ + 3^+4^+2
Z(s)
2s +5
(s + 2X* + 4)
Synthesize Z(s) = i
i\/ v ^ mto tne form
(s + IX* + 5)
s*

Z(s)

+2s*
5

11.14

r*Tnnp

in the figure.
6

If

shown

Zfr).

PROB.
11.15
sents

Of the

11.14

three pole-zero diagrams shown, pick the diagram that repre-

an R-L impedance function and synthesize in a

series Foster

JO)

-4

-3 -2

-l
ju

-4

-3

-2

-1

JO)

-4 -3 -2 -1

PROB.

11.15

10

form.

Network

340

analysis

and synthesis

11.16 Synthesize a driving-point impedance with the pole-zero pattern shown

in the figure in any form you choose. (Hint: Use uniform loading concepts.)

JU

*i
LU
PROB.
11.17

Following are four successive approximations of tanh

3s
(a)

s*+3

+ 105*
+ 455* + 105

IPs*
(c)

s*

11.16

s.

+ 15s
+ 15
s
s* + 1055 + 945*
W) 15** + 4205s + 945
s*

(*)

6s*

Synthesize networks for the functions above whose input impedances approxi-

mate tanh*.

chapter 12

Elements of transfer
function synthesis

12.1

PROPERTIES OF TRANSFER FUNCTIONS

A transfer function is a function which relates the current or voltage

another port. In Chapter ? we

at one port to the current or voltage at
terms of the opendiscussed various descriptions of two-port networks in
parameters y u . Recall that for
circuit parameters z and the short-circuit
open-circuit transfer imthe two-port network given in Fig. 12.1, the
pedances !, and Zu were defined as

h l/i-o

(12.1)

*1 lli-0

voltage-ratio transfer
In terms of the open-circuit transfer impedances, the

function

is

given as

T,

Yl

(12.2)

ratio
In terms of the short-circuit parameters, the voltage

Yl

is

= _ 2

z"

of the overall

= _?S*_
h
341

to be

(12.3)

When the network is terminated at port two by a resistor R,

Fig. 12.2, the transfer impedance

shown

+*

network

as

shown

in

is

(12.4)

Network

342

and synthesis

analysis

Two-port
network

Vi

V2

FIG.

The

FIG. 12.2

12.1

transfer admittance of the overall structure in Fig. 12.2

= -' =
Vi

where

G=

l/R.

y2

is

(12.5)

+G

When both ports are terminated in resistors,

Fig. 12.3, the voltage-ratio transfer function

shown in

z ai2

*a

V.

as

V^Vg is

(z

RMtn.

*)

(12.6)

- 2gii8

Other transfer functions such as current-ratio transfer functions can

and short-circuit parameters. In
Chapter 10 we discussed the various properties of driving-point impedances such as z u and z 28 This chapter deals with the properties of the
transfer immittances z M and y tl for a
also be described in terms of the open-

passive reciprocal network.

us

discuss

certain

First, let

properties

which

apply to all transfer functions of passive

linear networks with lumped elements.
We denote a transfer function as T(s).
1.

FIG. 12.3

T(s) is real for real s. This property

satisfied

is

when

T(s)

is

a rational

function with real coefficients.

2. T(s)

has no poles in the right-half plane and no multiple poles on

If T(s) is given as T(s)

P(s)lQ(s), the degree of P(s) cannot
exceed the degree of Q(s) by more than unity. In addition, Q(s) must be

they'to axis.

a Hurwitz polynomial.
3.

that

Suppose P(s) and Q(s) are given in terms of even and odd
is,

ns)
where Ms)

is

m = MMIM
m +

parts,

(12 7)
.

w,(5)
e(S)
8( S )
even and Ns) is odd. Then T(ja>) is

MiC/oO
MJja>)

The amplitude response of

\T(j<o)\

+
+

A^
NMa>)

T(ja>) is

=
UfA/co)

+ Nt*(ja>)]

(12.9)

Elements of transfer function synthesis

and

is

an even function in m. The phase response

Arg TQm)

is

mm

= arctan

343

'

0, we see that the phase response is an odd function in <o.

us discuss some specific properties of the open-circuit and
short-circuit parameters.

If arg T(jO)

Now

let

1. The poles of zgl (,s) are also the poles of z (j) and z {s). However,
u
it
not all the poles of z u (s) and z M (y) are the poles ofzn (s). Recall that in
Chapter 9 we defined the z parameters in terms of a set of node equations

as

Z^s)

Zl ^12

2.o

is no cancellation between each numerator and denominator of z

u,
and z lg then the poles are the roots of the determinant A, and all three
functions have the same poles. Consider the two-port network described
by the black box in Fig. 12.4a. Let z' u z' a2 and z' lt be the z parameters
of the network. Let us examine the case when we attach the impedances
Zx and Zg to ports one and two, as shown in Fig. 12.46. The z parameters

If there

zlt ,

for the two-port network in Fig. 12.46 are

= z ii + Zj
Z 22 Z 22 + Zj
Z 12 = Z 12
z ii

poles of z u include the poles of Z^ the poles of z

M
include the poles of Zj. However, the poles of z
lt include neither the
poles of Zi nor Z,. Consequently, we see that all the poles of z are also
It is clear that the

poles of z u

and z. The

h
+

l_
*\l> 2*22
z'tt

reverse

is

not necessarily

true.

I,

v2

r
Vi

<*'ll. *'22

'l2

(a)

FIG. 12.4

3j

-o

v2

Network

344

anal/sis

and synthesis

A-i

r^

Vi

Y!

4"

y'n, y'22
y'12

"

3f

/2

>

Y2

V2

^T

*h

Vi

V*

FIG. 12.6

FIG. 1X5

poles of yis(s) are also the poles of y u(s) and y M(s). However,
not all of the poles of yu(s) and y M(s) are the poles of yM(s). This property
is readily seen when we examine the two-port network in Fig. 12.5.
2.

The

The y parameters are

= y'* +
yit = v'm

y**

y%

y12(s) do not include the poles of either

Consider the network in Fig. 12.6. The y parameters are

Clearly, the poles of

2
ViM = -s +
y(s)

Yt

and

Yt

3s

= - + 3s
s

yis(s)

-3s

and s = 00, whereas

at s =
Observe that yu(s)
=
00.
s
pole
at
has
a
ylt(s) only
Let us
3. Suppose yu (s), y M(s), and y 18(s) all have poles at s = st
denote by u the residue of the pole at st of the function y u (s). The
residue of the pole s = Sj, of y w(s) will be denoted as km , and the residue
of the same pole of yia (s) will be denoted as k lt Without going into the
proof, 1 a general property of L-C, R-C, or R-L two-port networks is that
and yM(s) have poles

Jfc

k ak M

(12.11)

as the residue condition. For example, for the

network in Fig. 12.6, the residue condition applied to the pole at
3*
0, we have
0; whereas for the pole at s
00 gives 3 x 3
is fulfilled
condition
0*
residue
the
see
that
we
Thus
0.
8

This equation

L-C

- klt ^

2x4

is

known

= >

for both poles.

l
For a general discussion, see M. E. Van Valkenburg, Introduction
Network Synthesis, John Wiley and Sons, New York, 1960, pp. 305-313.

to

Modem

Elements of transfer function synthesis

345

/2

FIG. 12.7

-T7>

Vi

h
<

+
Vj

FIG. 12.8

111

ZEROS OF TRANSMISSION
zero of transmission

transmission, there

is

is a zero of a transfer function. At a zero of

zero output for an input of the same frequency.

For the network in Fig. 12.7, the capacitor is an open circuit at s = 0,

so there is a zero of transmission at s = 0. For the networks in Figs.

12.8 and 12.9, the zero of transmission occurs at s

//VZc. For
the network in Fig. 12.10, the zero of transmission occurs at s
l/RC.
In general, all the transfer functions of a given network have the same

zeros of transmission, except in certain special cases. For example if

(*) has a zero of transmission at s
su than y(j), V^lV^s), etc.,

h
k

v2

FIG. IX*

FIG. 12.1*

346

Network

analysis and synthesis

*1

z3

Zx

y2

Vi

ZB

Y<1

FIG.

^6

12.11

s^ This fact is clearly seen when we examine

have a zero at s
the relationships between the transfer functions. For example, we have

will also

Zi

yn

yuVn.

and

(12.12)

Vi^ti

Vn-

(12.13)

In addition, the voltage- and current-ratio transfer functions can be

expressed in terms of the z and y parameters as

Yl

5ll

h Usk

(12.14)

In Chapter 8 we saw that transfer functions that have zeros of transmission only on the jco axis or in the left-half plane are called minimum
phase functions. If the function has one or more zeros in the right-half
plane, then the function is nonmimimum phase. It will be shown now that

any transfer function of a passive reciprocal ladder network must be

minimum phase. Consider the ladder network in Fig. 12.11. The zeros of
transmission of the ladder occur at the poles of the series branch impedances or at the zeros of the shunt branch impedances. Since these
branch impedances are themselves positive real, the poles and zeros of
these impedances cannot be in the right-half plane. Consequently, the
transfer functions of ladder networks must be minimum phase. For the
network in Fig. 12.12, a transfer function would have two zeros of

FIG. 12.12

Elements of transfer function synthesis

347

FIG. 12.13

due to the elements Lx and C8 It would also have a

= due to Clt a zero at s = l/Ra Ct due to
the parallel R-C branch, and a zero of transmission at s = jULtCtf*
due to the L-C tank circuit. It is seen that none of the transmission
zeros are in the right-half plane. We also see that a transfer function may
transmission at s

oo

zero of transmission at s

possess multiple zeros

on

the ja> axis.

In Section 12.4 it may be seen that lattice and bridge circuits can easily
be nonminimum phase. It can also be demonstrated that when two ladder
networks are connected in parallel, the resulting structure may have
right-half-plane zeros. 2

12.3

SYNTHESIS OF

AND Zn WITH A

I-S2

TERMINATION

In this section we consider the synthesis of an L-C ladder network

with a 1-Q resistive termination to meet a specified transfer impedance Zji
or transfer admittance Ytl . In terms of the open- and short-circuit

parameters of the

L-C circuit, Z^C?) can be

Zm
and

Ytl(s) is

'21

I"*-'

Before

and

(12.15)
1

Vn

y
as depicted in Figs. 12.13

expressed as

(12.16)
i

12.14.

we proceed with the actual details of the synthesis,

it is

necessary

two important points. The first deals with the ratio of the odd
to even or even to odd parts of a Hurwitz polynomial Q(s). Suppose
to discuss

FIG. 12.14

Van Valkenburg,

op.

cit.,

Chapter

11.

Network

348
Q(s)

analysis and synthesis

given as

is

Q(s)

where M(s)

is

M{s)

the even part of Q(s),

+ N(s)

and N{s)

(12.17)

is

the

odd

part.

We know

that the continued fraction expansion of M(s)lN(s) or N(s)IM(s) should

yield all positive quotients. These quotients can, in turn, be associated
with reactances. Therefore it is clear that the ratio of the even to odd or

the

odd

to even parts of a Hurwitz polynomial

is

an L-C driving-point

function.

The second point to be discussed is the fact that the open-circuit transfer
impedance ztl or the short-circuit transfer admittance yn of an L-C circuit
that in an L-C
is an odd function. To show this, we must remember
90 out of phase with the
circuit with steady-state input, the currents are
voltages.

Thus the phase

voltages or input voltages

shifts between the input currents and output

and output currents must be 90 out of phase,

or

and

Arctan^=Jrad

(12.18)

Arctan^=?rad

(12.19)

Re j/ 21(;a)) for an L-C network. In order for the

so that Re z n (jw)
of an L-C two-port
real parts to be equal to zero, the functions z and y ai
network must be odd.
Suppose, now, that the transfer admittance
of two polynomials*

Y11 =
where P(s)
circuit

is

either

^ = ^+
M{s)

Q(s)

even or odd.

Y21 is given as the quotient

(12.20)

N(s)

Now, how do we

determine the short-

parameters y n and y M from the Eq. 12.20 to get

ya

Y" =

i+y

it

into the

form

(12.21)

We divide both the numerator P(s) and the

N(s), the even or the odd part of Q{s).
or
denominator Q(s) by M(s)
even, we divide by N(s) so that
is
P(s)
must be odd, if
Since

The answer

is

quite simple.

ytl

P(s)/N(s)

In
1

[M(s)IN(s)]

(12>22)

Elements of transfer function synthesis

From this we

yn

obtain

349

(12.23)

_M(s)
y"

On the

other hand,

if P(s) is

JV(s)

odd,

we

divide

by M(s) so that

Z(!VMW_
+ [tf(s)/M(s)]

y
1

(12 24)
.

M(s)
(12.25)

yM

M(5)

We assume that P(s), Mis), and N(s) do not possess common roots. For
our purposes, we will consider only the synthesis of Yn or Ztl with zeros

oo. In a ladder network, a zero of

of transmission either at s = or s
= corresponds to a single capacitor in a series branch
or a single inductor in a shunt branch. On the other hand, a zero of
oo corresponds to an inductor in a series branch or
transmission at s
a capacitor in a shunt branch. In terms of the transfer impedance

transmission at s

Zn(s)

= fW =

*('"

-i"-

"i

>

(12 26)
.

implies that the coefficients an_ x ,

ax , a are all zero. The number of zeros of Z,^) at s = oo
is given by the difference between the highest powers of the denominator
and the numerator, m n. We know that n can exceed m by at most
unity, while m can be greater than n by more than one. For example,
, a lt a = 0, we know that the
if m n = 2, and n = 3 with _i,
and two zeros
transfer function has three zeros of transmission at s =

the presence of n zeros of Z, x (.y) at s

a_,

oo.
of transmission at s
We can now proceed with the matter of synthesis.
following example.

sr

We

3s

+ 4s + 2

see that all three zeros of transmission are at j

numerator P(s)

is

a constant,

it

Consider the

oo.

Since the

must be even, so we divide by the odd

350

Network

analysis

and synthesis

part of the denominator s*

+ 4s. We then obtain

a

+ 4s
g
3ss + 2
+ 4s
s

z as

(12.28)

We see that both zsl and z2g have the

same poles. Our task is thus simpliz 22 so that the resulting network
synthesize
must
where
we
fied to the point
requires
that we first examine the
z
This
of
zeros
transmission
has the
21
possible structures of the networks which have the required zeros of
transmission and see if we can synthesize z 22 in one of those forms. For
the example that we are considering, a network which gives us three zeros
of transmission at s = oo is shown in Fig. 12.15. We can synthesize z22
to give us this structure by the following continued fraction expansion
.

of

l/z 22 .

3j

2)i

4s(ls+-Y

s+

|s

^)3j*

2(&s +-Z

3*

2Ys(is^Y

from the l-2 termination toward the input end,

the final network takes the form shown in Fig. 12.16. Examining the
network more closely, we see that it takes the form of a low-pass filter.
Since z 22

is

synthesized

Thus the specification of all zeros at s

oo

is

equivalent to the specification

of a low-pass
As a second example, consider the transfer impedance
filter.

*>Since the numerator of

Zgl

(,s)

is

,+

<12S)
4.

an odd function, we have to divide both

-nRPP-

-ffflHT1

<D T
FIG. 12.15

+2

FIG. 12.16

Elements of transfer function synthesis

351

FIG. 12.17

numerator and denominator by the even part of the denominator so that

*n

-*

=
3s*

*
""

+2

s*

+ 4s
T
+2

(12.30)

3s*

at s =
is a high-pass
by a continued fraction expansion of M .

The network that gives three zeros of transmission

structure

The

which

is realized

final realization is

shown

in Fig. 12.17.

Finally, consider the transfer admittance

Yn(s)

+ 4s + 2
which has two zeros of transmission at s = 0, and one zero
Since the numerator is even, we divide by s* + 4s so that
3s' + 2
y = -rrr
s + 4s
s*

(12.31)

3s*

at s

oo.

~x

(12.32)

The question remains

as to how we synthesize y a to give a zero of transmission at s

oo and two zeros at s
0. First, remember that a parallel
inductor gives us a zero of transmission at s
0. We can remove this
parallel inductor by removing the pole at s
of y n to give

Vi

i/
Vu

2s

5s/2

= ~
s* + 4

(12.33)

If we invert ylt we see that we have a series L-C combination, which gives
us another transmission zero at s
0, as represented by the f-farad
capacitor, and we have the zero of transmission at s
oo also when we
remove the inductor of f h. The final realization is shown in Fig. 12.18.

H(2h<

)Vt

FIG. 12.18

352

Network

analysis

and synthesis

1Q<

Network

Zn

\)
I

(a)

An important point to note in this synthesis procedure is that we must

place the last element in series with a voltage source or in parallel with a
current source in order for the element to have any effect upon the transfer
function. If the last element

is

denoted as

Zn

then the proper connection

of Z n should be as shown in Fig. 12.19.

12.4

SYNTHESIS OF CONSTANT-RESISTANCE NETWORKS

In this section we will consider the synthesis of constant-resistance twoport networks. They derive their name from the fact that the impedance
looking in at either port is a constantresistance R when the other port is terminated in the same resistance R, as depicted
Two-port
network

in

Fig.

works

12.20.

Constant-resistance

net-

are particularly useful in transfer

because when two

networks with the
network.
same R are connected in tandem, as shown
loads down the other. As a result, if the
network
neither
in Fig. 12.21,
voltage-ratio transfer function of a is V^V-l and that of b is VJVS the
voltage-ratio transfer function of the total network is
function

FIG.

12.20. Constant-resistance

synthesis,

constant-resistance

Y*Y

(12.34)

Equation 12.34 implies that, if a voltage-ratio transfer function is to be

ratio could
realized in terms of constant-resistance networks, the voltage

Elements of transfer function synthesis

*.

t.

FIG.

Na

V2

353

R<

Nb

12.21. Constant-resistance

networks in tandem.

could be
be decomposed into a product of simpler voltage ratios, which
tandem.
in
connected
then
realized as constant-resistance networks, and

For example, suppose our

objective is to realize

- z)(s - zi)(s - z,)

- p )(s - Pi)( - P)

K_

K(s

Va

(s

We

in terms of constant-resistance networks.

individual voltage ratios

can

(12.35)

first

synthesize the

_ K (s - 2 )
s Po
V.

\\

v*

= K^s
s

Y*

VhjVa

p,
in

Although there are many

works,

and then connect the three networks

as constant-resistance networks
realize

(12.36)

Pi

-zj
= Xa(s

Vt
tandem to

-*i)

different types of constant-resistance net-

we will restrict ourselves to networks of the bridge- and lattice-type

structures as

shown

in Figs. 12.22o

and 12.226. These networks are

the input and output ports do not possess comclose examination, we see that the bridge and
a
Upon
terminals.
mon
are identical circuits. The bridge circuit is
12.22
Figs.
in
circuits
lattice
merely the unfolded version of the lattice. Consider the open-circuit

balanced structures;

i.e.,

parameters of the bridge circuit in Fig. 12.23.

impedance

First

we determine

the

as
ii

Next we determine the

transfer

(12.37)

impedance zn , which can be expressed as

= V* ~ V
h

*'

z ai

(12.38)

Network

354

analysis and synthesis

'.

(a)

+"

(b)

FIG.

and

is

12.22. (a) Bridge circuit,

obtained as follows.

We first obtain the current /' as

r_
Next we

(6) Constant-resistance lattice.

Za + Z b

Vt - Vt =

find that

_h

Vi

(Zb

(12.39)

- Z )I'
tt

= (Z -Za)|
6

FIG.

12.23. Analysis

of bridge

circuit.

(12.40)

Elements of transfer function synthesis

^ = Z> ~ Z

so that

From

Now

(12.41)

the lattice equivalent of the bridge circuit

us consider the

355

we

note that z M

= zu

terminated in a resistance
R, as shown in Fig. 12.226. What are the conditions on the open-circuit
parameters such that the lattice is a constant-resistance network? In
other words, what are the conditions upon z u and z n such that the input
impedance of the lattice terminated in the resistor R is also equal to R7
let

In Chapter 9

we found

lattice circuit that is

that the input impedance could be expressed as

Zu
Since z u

= *ii - -SjJ-r
Zn + R

= zu for a symmetrical network, we have

Z" "

In order for

Zu

z 11

lattice

network,

- zia =

(12.44)

we then have

iKZ,
which

(1243)

+*

= R, the following condition must hold.

*u*

For the

(12.42)

+ Z,)* - (Za - Z)] = **

(12.45)

R*

(12.46)

Z^Lb

simplifies to give

Therefore, in order for a lattice to be a constant-resistance network,

Eq. 12.46 must hold.

Next, let us examine the voltage ratio VjVg of a constant-resistance
lattice whose source and load impedances are equal to R (Fig. 12.24).
From Chapter 9 we can write

z^R

V9

(zu

+ K)(zM + R) - ztfr*

FIG. I2J4. Double-terminated

lattice.

(12.47)

Network

356

which

anal/sis

simplifies to

and synthesis

IS-

5*&ii

(zn

V.

+ Kf - **i

= __Zt__2
2Rz11 + 2R
In terms of the element values of the
V,

From

(12.48)

lattice,

we have

- Za)R
R(Z b + Za) + 2R*
l(Z b

(12.49)

the constant-resistance condition in Eq. 12.46,

V,

UZ b - (RVZh)]R

Vt

we obtain

R[Z b

(R*/Z)]

- R*)
+ R8) + 2RZ
a
a
_ j(Zt - R )
(Z + R?

2R a

KZ

(Z 6a

* (Z >

R)
(12.50)

Zb + R

In Eq. 12.50, the constant multiplier J comes about from the fact that
the source resistance

R acts as a voltage divider.

G(s)

we can

express

Z6 in Eq.

12.50 in terms of

HP +

If

we

let

(12.51)

G as

G(s)]

1-G(5)

"

In terms of Z, the voltage ratio can be given as

V2
Va

1R-Z

In the following examples, we will usually

Example

12.1.

The voltage

(12.53)

2R + Za
let

be normalized to unity.

ratio is given as

2l=lJlI
2s + 1
V

(12.54)

Elements of transfer function synthesis

we

which, as

recall, is

with Eq. 12.50,

an

all-pass transfer function.

By

associating Eq. 12.54

we have
(12.55)

Z =>s,
Since

Z^Za =

1,

we then

Za --s

obtain

(12.56)

We see that Z is a 1-h inductor and Za is a 1-farad capacitor.

is

shown

357

The final network

in Fig. 12.25.

If

FIG. 12.25

Example

12.2.

Let us synthesize the all-pass function

V ~2s +

'

whose pole-zero diagram

is

shown

l s

2s

V
let

First,

we

~
+

Since the portion

l
1

us concentrate on synthesizing the function

Vt

j*

-2s +2

s*

+2s+2

V~
separate the numerator

even parts. Thus

(12.57)

+ 2s+2

in Fig. 12.26.

Yl
has already been synthesized,

s*

(12.58)

and denominator function

into

odd and

we have
V*

Va

(f

(5*

+ 2) - Is
+ 2) + 2s

(12.59)

jut

-<T

yi

-1

+1

-fl

FIG. 123*

Network

358
If

we

analysis

and synthesis

divide both numerator

and denominator by the odd part

Vt

[(*
[(5*

We then see that

+ 2)/2t] + 2)/2*] +
a* + 2
s

2s,

we

(12.60)
1

=2 +I
which consists of a $-h inductor in
pedance Z is then

series with

*-JT5
and

is

obtain

(12.61)

1 -farad

capacitor.

The im-

(1262)

recognized as a J-farad capacitor in parallel with a 1-h inductor. The

VjVa is thus realized as shown in Fig. 12.27. The structure that

voltage ratio

FIG. 12.27

VJVg in Eq. 12.S7 is formed by connecting the

networks in Figs. 12.25 and 12.27 in tandem, as shown in Fig. 12.28. Finally,
it should be pointed out that constant-resistance lattices can be used to realize
other than all-pass networks.

realizes the transfer function

FIG. I2JS

Next,
12.29.

let

us consider the constant-resistance bridged-T network in Fig.

network are all equal to R ohms, the

If the resistances in the

network has constant-resistance

if

ZA - *

(12.63)

359

Elements of transfer function synthesis

ED
AMr
R

-vwR

Vi

.*.

Zb

Vt

FIG. VIM. Constant-resistance bridged-T

Under

circuit.

the constant-resistance assumption, the voltage-ratio transfer

function can be given as

V*

(12.64)

Zt + R

R + Za

Vx
Exanple 123.

z>

Let us synthesize the voltage ratio

V.

s*

j*+25 +

(12.65)
l

as a constant-resistance bridged-T network terminated in a 1-G resistor. First

let

us write

VJV1 as

V1
Vt

[2*/(*

(12.66)
1)]

Is

so that

Zb

and

We

recognize

circuit.

(12.67)

+l
~ s* + l
*

Za

as a parallel

The final network

is

L-C tank

shown

circuit

(12.68)

and Z as a

in Fig. 12.30.

2h

r-CH
-a/v\

10

i-vwk

10

10!
:2f

FIG. I2J0

series

L-C

tank

Network

360

Example

analysis

and synthesis

Let us synthesize the voltage ratio

12.4.

(s

Vs

Fx "
in terms of

At

first,

(5

+ 2X* + 4)
+ 3X3* + 4)

two constant-resistance bridged-T

we break up

circuits

(12.69)

connected in tandem.

the voltage ratio in Eq. 12.69 into two separate voltage

ratios

S+1
5+3

Va

Vi

+4

and

va

For the voltage

ratio

(12.70)

(12.71)

35+4

VJVi, we have
s

5+3

zbl +

(12.72)
1

so that

Zn

For the voltage

+ 2 and
ratio

zal ~5+2

vjva we have

5+4
35+4
AnH

(12.73)

1
1

+zot

(12.74)

25
jr_.

FIG.

5+4

12.31

(12.75)

Elements of transfer function synthesis

and

Zha

The final

synthesized network

is

shown

361

+4
(12.76)

'

2*

in Fig. 12.31.

Problems
12.1 Give an example of a network where: (a) a transfer function has
multiple zeros on they'to axis; (b) the residue of a pole of a transfer function on
thsjo> axis is negative.

12.2

Show

that the residue condition holds for the networks

shown

in the

figure.

lo-

lh
'TflHP

3h

10

lo-^VVW-

-o2

af

-o2

Hf:io

=fc*f

iq:

-o2'

l'o-

-o2'

l'o(b)

(a)

PROB.

123 For the network shown,

and plot on a complex plane.

find

12.2

by inspection the zeros of transmission

nnnp
ff

lf^=

^=2f

For the networks in the figure

are equal to
when ZZt
R*.

12.4
ln

.20

f:

lh

S10

PROB.

in.

12.3

show

that the driving point impedances

362

Network

anal/sis

and synthesis

s
R

JC

Vi Zfci-*

zb

* *

(a)

%
Vi

Zta -ii

*^

PROS.
123 For the networks in Prob.

v2

12.4

12.4, find the voltage-ratio transfer functions

yjvv
12.6

For the network shown

in the figure (a)

Yl.
V
(b) Synthesize

2+Y

+ 2)
+ 2s + 2

O.S(*

s*

10

PROB.
(a)

that

Ywhen
l\

12.7

show

<

V2

12.4

For the constant-resistance bridged-T circuit, show that if Z

Zh 1,

tt

then

L_

Vt

+za

Elements of transfer function synthesis

(b) Synthesize

Za and Zh if

_
_
V ~
l

12.8

363

s*
s*

+3s +

+45* +5j

+2

Synthesize the following voltage ratios in one of the forms of the

networks in Prob. 12.4

<)

V,

s+2

Vl

Vt

Vx " 2** + 2* + 6
3(jr + 0.5)
Vt
Vx " 4j + 1.5

(e)

12

2(s*+3)

Synthesize iV with termination resistors /tt

=4fi,i?1

~into

give

12j*

Yl

"

V,

15i

+ 7* + 2

PROB. IL9
12.10

For the network

in Prob. 12.9, realize network

X

Yl
(a) Synthesize AT. as

_L_
+3

2 2s

Na to give

a constant-resistance

lattice. (R =- 1 n.)
a constant-resistance ladder as in Prob. 12.4.
as a constant-resistance bridged-T circuit. (R -

(6) Synthesize AT. as

(c) Synthesize A',

12.11

()

Synthesize the following functions into the form

s*

(*)

(c)

id)

Z- *

+3s* +3s

+2

3s*

3s

3s

s*

'a

s*

s*

+ 3s* + 3s +2

(*

+ 2)

3s*

+
s*

Y-

s*

(e)
'

shown

(Jt
1

Q.)

Q.)

in the figure

364

Network

anal/sis

and synthesis

PROB.I2.II
12.12

()

(b)

12.13
circuits.

Synthesize as a constant-resistance lattice terminated in a 1-Q resistor.

-s+1
+s + 1
h ~ s* - 3s + 2
vt s* + 3s +2
via s* - 20s* + 5s Vx(s) ^ s* + 20s* + 5s +
V,

s*

Vx

s*

20
20

Synthesize the functions in Prob. 12.8 as constant-resistance bridged-T

chapter 13

Topics

13.1

THE

FILTER DESIGN

design

in filter

PROBLEM

In the preceding chapters we examined different methods for synthesizing

a driving point or transfer function H(s). Most problems have as their
initial specification an amplitude or phase characteristic, or an impulse
response characteristic instead of the system function H(s). Our problem
is to obtain a realizable system function from the given amplitude or
phase characteristic. For example, a typical design problem might be to
synthesize a network to meet a given low-pass filter characteristic. The
specifications might consist of the cutoff frequency m c the maximum
allowed deviation from a prescribed amplitude within the pass band, and
the rate of fall off in the stop band. We must then construct the system
function from the amplitude specification. After we obtain H(s), we
proceed with the actual synthesis as described in the Chapter 12. Another
problem might consist of designing a low-pass filter with a linear phase
characteristic within the pass band. Here, both amplitude and phase are
specified. We must construct H(s) to meet both specifications. Problems
of this nature fall within the domain of approximation theory. In this
chapter we will consider selected topics in approximation theory and then
present examples of filter design where both the approximation and the
synthesis problems must be solved.
,

13.2

THE APPROXIMATION PROBLEM

NETWORK THEORY

The

IN

essence of the problem

is the approximation of a given funcby another function/^t; !,..., a) in an interval xx ^ x ^ xt

The parameters ax
in the approximating function are fixed
, <x

tion/(rr)

365

Network

366

by the
/at*;

analysis and s/nthesis

<*i

*n) the following error criteria are

Least squares. The value of /(alf

1.

When we

particular error criterion chosen.

and w(x)

let

= f{x)

most common:

aj is minimized where

a weighting function which stresses the error in certain sub-

is

intervals.

Maximally flat. The

2.

at

first

derivatives of e are

made

to vanish

= *o.

x
3.

Chebyshev. The value of p

where

/*

is

minimized in the interval *i<,x <,xt

|lnx-

The value of
<, xt

4. Interpolation.

in the interval x^x

e is

made

to vanish at a set of n points

is chosen, we must determine the particular

function. This depends upon whether we
approximating
form of the
time or frequency domain. Suppose /(a;)
the
in
approximate
choose to
represents a magnitude function in the frequency domain and the approximating function is to be rational in co*; then

After an error criterion

/(*;

!,..,*)=

(13.1)

where * = to*. In addition, the values of xk must be restricted to insure

^ 0. In the time domain, /(&) might represent an
that/ (x; alf
,
system to be synthesized. In the case of an R-C
of
a
impulse response
we
have
function,
transfer
.

fa(x; 04,
where x

t.

Since an

...

oc)

**'

oe,

e"*

(13.2)

R-C transfer function must have its poles on the

<x t , k even, are restricted to negative real

negative real axis, the values of

numbers.
The keystone of any approximation problem

lies in

the choice of a

suitable error criterion subject to realizability restrictions.

can be simplified

when some of the

oc's

The problem

are assigned before applying the

error criteria. All the error criteria cited, except the Chebyshev, can then
be reduced to a set of linear algebraic equations for the unknowns
ii

>

Time-domain approximation
The principal problem of time-domain approximation

consists of

approximating an impulse response h(0 by an approximating function

Topics

in filter

367

design

h*(t) such that the squared error

= ["[fc(0-**(')f *

minimum.

is

A generally effective procedure in time-domain approximation utilizes

orthonormal functions ^(r). 1 The approximating function h*{t) takes
the form

**(0-Z.0
k=l

(13.3)

so that the error

Jo

*&(<)] dt

*-i

minimized when

is

.=r

h(t)<f>k (t)dt

l,2,...,n

(13.4)

we saw in Chapter 3. If the orthonormal set is made up of a

of exponentials e**', then the approximate impulse response
as

,ri

sum

(13.4)

*=1

#*(*)=!

has a transform

*=1 s
Realizability is insured if in the

1, 2,

n.

(13.5)
sk

orthonormal

Synthesis then proceeds

set {a te***},

Re ^

^ 0;

from the system function H*(s)

obtained in Eq. 13.S.

Frequency-domain approximation
In frequency-domain approximation
the

principal

problem is to find a
whose magnitude

rational function H(s)

|/f(/Yu)|

approximates the ideal low-pass

-1

+1

characteristic in Fig. 13.1 according to

a predetermined error criterion. In the

next few sections we examine several

W. H. Kautz, "Transient Synthesis in the

Theory, CT-1, No. 3, September 1954, 29-39.
1

FIG . I3X

Ideal

fihcrchar_

acteristic.

Time Domain," IRE

Trans, on Circuit

Network

368

different

analysis

and synthesis

ways to approximate the

the maximally flat or

ideal low-pass:

Butterworth approximation, the equal-ripple or Chebyshev approximation,

and the optimal or Legendre approximation. Another major problem is
that of obtaining a transfer function H^s),

whose phase

is

approximately

approximately flat over a given range of frequencies.

Here again, there are two different methods the maximally flat or the
equal-ripple methods. Our discussion will center around the maximally
flat method. The joint problem of approximating both magnitude and
linear or whose delay

is

phase over a given frequency range

is

possible, but will not be discussed

here.

13.3

THE MAXIMALLY FLAT LOW-PASS

APPROXIMATION

In Chapter 10

we saw

FILTER

that the ideal low-pass

filter

in Fig. 13.1

is

not

not zero for t < 0.

However, if we use a rational function approximation to this low-pass
filter characteristic, the Paley- Wiener criterion will be automatically
satisfied. We will therefore restrict ourselves to rational function approxirealizable because

its

associated impulse response

is

mations.

In low-p.ass filter design, if we assume that all the zeros of the system
function are at infinity, the magnitude function takes the general form

where

is

the d-c gain constant and/(<u 2)

to give the desired amplitude response.

/()

is

the polynomial to be selected

For example,
a)

if

2"

(13.7)

then the amplitude function can be written as

M()

,.

(1

We see that M(0) =

eo.

K,,,

+ o>\2nn)^

and that M(a>)

is

monotonically decreasing with

In addition, the 0.707 or 3-decibel point

M(l)

= -^

< 13 - 8 )

is

alln

at to

for

all n,

that

is,

(13.9)

is thus seen to be <w = 1. The parameter n conof approximation in both the pass band and the stop
band. Curves of M(m) for different n are shown in Fig. 13.2. Observe

The

cutoff frequency

trols the closeness

Topics
Radian frequency,
0.3

0.2

0.1

0.6

0.4

369

design

In filter

a
0.8

5^ ,v

-2

-4
\

-6
n=3

-8

n=5
n

\VV

=7

-10

I*

-12

\\\

-14

^ \

-16

\\

-18
-20

\\

-22

-24

-26
FIG.

13.2.

Amplitude response of Butterworth low-pass

filters.

that the higher n is, the better the approximation. The amplitude approximation of the type in Eq. 13.8 is called a Butterworth or maximally flat
response. The reason for the term "maximally fiat" is that when we
expand M(to) in a power series about <o = 0, we have
Jlf(eo)

We
(0

- Ao(l -

l<o

see that the first 2n

= 0.

For

>

1,

tn

+ |co 1

A<o"

i^eo""

(13.10)

derivatives of M{o>) are equal to zero at

the amplitude response of a Butterworth function

can be written as (with Kg normalized to be unity)

M(w) =!

<wl

(13.11)

to"

We observe that asymptotically,

Af(to) falls off as

wrn

response. In terms of decibels, the asymptotic slope

20 log M(co)

= 20/i log <w

is

for a Butterworth

obtained as
(13.12)

Network

370

analysis

and synthesis

Consequently, the amplitude response

6m db/octave or 20n db/decade.

One

How

question remains.

from only the amplitude

follows.

falls

asymptotically at a rate of

do we obtain a

transfer function H(s)

characteristics M(co)l

The procedure

is

as

We first note that the amplitude response M{m) and the complex

system function H(jco) are related by

M\w) =
If

we

new

define a

(13.13)

function A(s 2) such that

A(s 2)

we

H(j<o) H(-jco)

see that

H(-s)

H(s)

(<o)

(13.14)

^(-ot) 2)

(13.15)

From A( eo 2) all we need do is to substitute s* = to2

Then we factor A(s2) into the product H(s) H(s). Since

to give (s 2).

the poles

and

zeros of H(s) are the mirror images of the poles and zeros of H(s), we
simply choose the Hurwitz factors of A(js) as H(s). An example will serve

Consider the third-order (n

to clarify this discussion.

3)

Butterworth

response given by
JW

\OJ)

+ to

(13.16)

"We see that AC?2)

(s )

is

=
1

Factoring

(**),

(13.17)

(13.18)

--(s8)8

we obtain
1

(**)

2s

2s

s'

2s

2s

s*

= H(s)H(-s)

(13.19)

We then have
if(s)

=
s

2s

2s

1
1

l)(s

+ i + A/3/2X* + i - A/3/2)

(13-20)

The poles of #(y) and fT(-*) are shown in Fig. 13.3. Observe that the
poles of H(s) are mirror images of the poles of H(s), as given by the
theorem on Hurwitz polynomials in the Chapter 2.

Topics

13.3.

Poles of H(s)

design

371

-^ f->

HM^.-

FIG.

in filter

H(-s)

for

an n

= 3 Butterworth filter.

For a Butterworth response, the poles of H(s) H(s) are the roots of

(-1)V B = -1

=
The poles

**-"

or simply by

sk

Expressing st as s k

=
=

e*<-u/ft*
e

(13.21)
1

(13.22)

n odd
fc

ak

In

even

i(i/n) '

<+'-i>/an] I

0, 1,

by

s k are then given

sk

+ ja> k

= o, 1, 2,

,2n

(13.23)

the real and imaginary parts are given

by
ow

= cos 2fc + nn-

1
ir

sin

2n

/2fc

1\tt

/2
1

(13.24)
ct)v

2k
sin

+ n-

l
IT

cos

In

/2fc-

1\tt

12

It is seen from Eqs. 13.22 and 13.23 that all the poles of H(s) H(s) are
located on the unit circle in the s plane, and are symmetrical about both
the a and the jco axes. To satisfy readability conditions, we associate the
poles in the right-half plane with H(s), and the poles in the left-half

plane with H(s).

As an example, consider the construction of an H(s) that gives an n
4
Butterworth response. From Eq. 13.23, it is seen that the poles are
given by

st

^(fc+*)/8]

e"

(13.25)

Network

372
H(s)

anal/sis and synthesis

then given as

is

H($)

=
(S

If

we

i{

e ')(s

13.2 for n

(,

e>

*>*)

_
2

0.76536s

l)(s*

to n

8,

1.84776s

simplify the use of Butterworth functions, H(s)

and

(13 26)
"

and expand, we obtain

(s

To

*Xa

express sk in complex form

H( s\

13.1

.)(,

in factored

(12 27)
1)

given in Tables

is

form as in Eq.

13.27, or

multiplied out as

H(s)

-^
a ns

fln-xs"-

TABLE

a ts

(13.28)
1

13.1

Butterworth Polynomials (Factored Form)

+1
+ V2s + 1
(**+* + 1)(* + 1)
(s2 + 0.76536* + IX** + 1.84776* + 1)
8
2
(* + 1)(* + 0.6180* + IX* + 1.6180* + 1)
2
(** + 0.5176* + IX* + ^2* + IX*2 + 1.9318* + 1)
2
2
2
(* + IX* + 0.4450* + IX* + 1.2456* + IX* + 1.8022* + 1)
(s2 + 0.3986s + IX*2 + 1.1110* + IX*2 + 1.6630* + IX*2 + 1.9622* +

5s

4
5

6
7

TABLE

1)

13.2

Butterworth Polynomials *

a*

V2

2.613

3.414

2.613

3.236

5.236

5.236

3.236

6
7

3.864

7.464

9.141

7.464

3.864

4.494

10.103

14.606

14.606

10.103

4.494

5.126

13.138

21.848

25.691

21.848

13.138

1
1
1

5.126

Topics

13.4

OTHER LOW-PASS

In Section 13.3,
low-pass

filter

we examined

characteristic.

approximants in

FILTER

design

373

APPROXIMATIONS

the maximally

We

in filter

approximation to a

flat

will consider other low-pass filter

this section.

equal-ripple approximation
have seen that the maximally flat approximation to the ideal lowpass filter is best at co
0, whereas, as we approach the cutoff frequency
We now
<w = l, the approximation becomes progressively poorer.
consider an approximation which "ripples" about unity in the pass band
and falls off rapidly beyond the cutoff co = 1. The approximation is

The Chebyshev or

We

equally

good

at

co

ripple approximation.

and co = 1, and, as a result is called an equalThe equal-ripple property is brought about by the

use of Chebyshev cosine polynomials defined as

= cos (n cos-1 co)

= cosh (n cosh-1 co)
C^co) = 1
For n = we see that
and for n = 1, we have
Ci(<*>) = <
C(a>)

|co|

<, 1

|co|

>

(13.29)

(13.30)
(13.31)

Higher order Chebyshev polynomials are obtained through the recursive

formula
_
(B 32)
CJfu) =

^ ^^ ^ ^

Thus

for n

In Table

= 2, we obtain Cg(co) as
Cg(co) = 2<o((o)
= 2eo 8 - 1

13.3,

Chebyshev polynomials of orders up to n

TABLE
Chebyshev polynomials

(13.33)

10 are given.

13.3

Cn(m) = cos (n cos-1 <o)

1
1

CO

2co*
4e3

4
5

6
7
8

9
10

-1
3o>
8a>* - 8a>* + 1
16a> 6 - 20a>8 + 5o>
32o>* - 48<o + 18a>* - 1
64o> 7 - 112a> + 56eo8 - lea
128a>8 - 256w + 160w4 - 32o>* + 1
256o> - 576o7 + 432o) - 120a.8 + 9a>
512a 10 - 1280w8 + 1120a>* - 400eo* + 50* -

Network

374

FIG.

The

and synthesis

analysis

13.4. C,(co)

and

Chebyshev polynomials.

C,(u>)

pertinent properties of Chebyshev polynomials used in the low-pass

filter

1.

approximation are:

The

zeros of the polynomials are located in the interval

as seen by the plots of

2.

Within the interval

exceeds unity; that

3.

C3(o))

Beyond

values of

is,

and

the absolute value of

\w\ <, 1,

|C(eo)| <,

the interval

<

\a>\

|eo|

<, 1,

CJio)) in Fig. 13.4.

for

1,

\co\

CB (co)

never

<, 1.

|C(w)| increases rapidly for increasing

\a>\.

Now, how do we apply

the Chebyshev polynomials to the low-pass

approximation? Consider the function e 2 C%w), where e is real and
and e2
small compared to 1. It is clear that e 2 Cn 2 ((o) will vary between

filter

in the interval
e2

\u>\

Now we

1.

C 2(g>). This new function

slightly greater

than unity, for

we

the function which

add
<,

|<w|

will associate

IHO"))!*

=
1

<

to this function

varies between
1.

and

making
e2,

it 1

a quantity

Inverting this function,

we obtain

with \H(jco)\ 2 ; thus

+ *C n\a>)

(13.34)

\H(ja>)\ 2 oscillates

about unity such that the

Outside this interval,
Cn2(w) becomes very large so that, as to increases, a point will be reached
2
approaches zero very rapidly with
where e2 CB2(e>)
1 and \H(jcu)\
further increase in <o. Thus, we see that \H(jw)\ 2 in Eq. 13.34 is indeed a
Within the interval

maximum value is

|eo|

1,

and the minimum

is 1/(1

e 2).

approximant for the ideal low-pass filter characteristic.

Figure 13.5 shows a Chebyshev approximation to the ideal low-pass
<u <, 1, \H{jw)\ ripples
filter.
We see that within the pass band
suitable

between the value

and (1

e 2 )- -*.

The ripple height or distance between

Topics

(l

in filter

-^^-

+ e2) *

FIG.

3.5. Chebyshev approximation to low-pass

maximum and minimum

in the pass

<o

because

\H(jl)\

1, |H(;>)| is

Cn\\) =

where

e*

is

given as
1

(13.35)

(!+>"

+ ^a

Cn*{(o)

(13.36)

is,

for

\<o\

1,

as a> increases,

we

reach a point

so that

]ff(Mlas-

>

O)

eC n(a>)

The

filter.

1.

In the stop band, that

io k ,

band

=1

Ripple

At

375

design

(13.37)

0) k

loss in decibels is given as

Loss

But for large

<w,

CB(co)

= -20 log10 H(jm

^ 20 log e + 20 log CB(o)
\

)\

can be approximated by

its

(13.38)

_1
leading term 2" <w",

so that

Loss

= 20 log6 e + 20 log5 2*- g>"

= 20 log e + 6(n - 1) + 20/t log a>
1

(13.39)

Network

376

We

anal/sis and synthesis

Chebyshev response also falls off at the rate of 20/i

initial drop of 20 log e + 6(n 1) decibels. However,
in most applications, e is a very small number so that the 20 log e term is
see that the

db/decade after an

actually negative. It is necessary, therefore,

to compensate for this decrease in loss in

band by choosing a

the stop
cosh 0k

sufficiently

large n.

From

the preceding discussion,

we

see

a Chebyshev approximation depends

upon two variables, e and n, which can be
determined from the specifications directly.
The maximum permissible ripple puts a
bound on e. Once e is determined, any
that

sinh

0*

desired value of attenuation in the stop

band fixes n.
The derivation of the system function
H(s) from a Chebyshev amplitude approximation \H(jco)\ is somewhat involved and
not be given here. 2 Instead, we will
simply give the results of such a derivation.
will

First

we

introduce a design parameter.

FIG. 13.6. Locus of poles of

Chebshev filter.
fih

- sinh

where n

is

(13.40)

and e is the factor

The poles, sk = ak + jcok of the equal-ripple
located on an ellipse in the s plane, given by

the degree of the Chebyshev polynomial

controlling ripple width.

approximant H(s) are

sinh

cok

a
/S fc

cosh*

(13.41)

/?*

The major semiaxis of the ellipse is on the jco axis and has a value co =
cosh pk The minor semiaxis has a value a = sirih(ik and the foci
are at co = 1 (Fig. 13.6). The half-power point of the equal-ripple
.

amplitude response occurs at the point where the ellipse intersects the^eo
cosh fik . Recall that for the Butterworth response, the
axis, i.e., at co
half-power point occurs at co = 1. Let us normalize the Chebyshev poles
1 instead of at co
sk such that the half-power point also falls at co
cosh fi k ; i.e., let us choose a normalizing factor, cosh P k , such that the

Interested parties are referred to

Network

Synthesis,

M.

John Wiley and Sons,

E.

Van Valkenburg,

New

Introduction to

York, 1960, Chapter

13.

Modern

Topics

normalized pole locations

s' t

in filter

377

design

are given by

cosh

(th

cosh

/?*

cosh

= <**+./>*
locations can be derived as

The normalized pole

= tanhu &
sin

<r't

/2fc

- lW

^j
(13.43)

/2fc-lW

a>t

= cos ^__j_

Comparing the normalized Chebyshev pole locations with the Butterworth

we see that the imaginary parts are the same,

while the real part a' k of the Chebyshev pole location is equal to the real
part of the Butterworth poles times the factor tanh /3. For example, with
0.444, the Butterworth poles are
n
3 and tanh /?,.
pole locations in Eq. 13.24,

*i- -1 +J0
j23

_0.5 /0.866

so that the normalized Chebyshev poles are given by

4 --1(0.444)+ JO

szs

-0.444

-0.5(0.444) ;0.866

-0.222 y0.866

Finally, to obtain the denormalized

s' k

by cosh

/3 t that
,

+;0

Chebyshev

poles,

we simply

multiply

is,

Sk

(a\

+ /',) cosh &

(1 3.44)

There is an easier geometrical method to obtain the Chebyshev poles,

given only the semiaxis information and the degree n. First we draw two
circles, the smaller of radius sinh /?* and the larger of radius cosh /? as

Network

378

and synthesis

anal/sis

Butterworth
pole locus

Chebyshev
pole locus

FIG.

shown in

13.7.

Chebyshev

poles.

radial lines according to the angles of

we draw

Fig. 13.7. Next,

filter

the Butterworth poles (Eq. 13.22) as shown.

Finally,

we draw

vertical

dashed lines from the intersections of the smaller circle and the radial lines,
and horizontal dashed lines from the intersections of the large circle and
the radial lines. The Chebyshev poles are located at the intersection of
the vertical and horizontal dashed lines, as shown in Fig. 13.7.
Consider the following example. We would like to obtain a system
function H(s) that exhibits a Chebyshev characteristic with not more than
ripple in the pass

1 -decibel

When we design for

down

band and is down

1-decibel ripple,

\H(jl)\

20 log -t-zu

'(!+)*

We then obtain
(1

and

* )
e

Our next

task

is

to

1,

Solving for ,

1A

= -1

w=

2.

\H(j\)\ is

0.509

^ 20 log 0.509 + 6(n =

134 5)

(13.46)

(13.47)

from the 20 decibels at m = 2

can be given as approximately

we obtain

0.891

to find n

Eq. 13.39 the loss

20

1)

specification.

20 log 2

(13.48)

n must be an integer, we let n = 3.

the pole locations are completely speci-

2.65. Since

With the specification of n and e,

Our next task is to determine

fied.

20 decibels at

that at

decibel so that

20 log

From

at least

we know

these pole locations.

First

we must

Topics
find

Pk

From

Eq. 13.40

in filter

379

design

we have

= I sinh"

(13.49)

1.965

0.476

In order to find the normalized Chebyshev poles from the Butterworth

poles, we must first determine tanh fik Here we have
.

tanh

From Table

13.1, the
*!

pk

tanh 0.476

= 0.443

(13.50)

3 Butterworth poles are

-1.0,

s 23

-0.5 y0.866

(13.51)

Multiplying the real parts of these poles by 0.443, we obtain the normalized

Chebyshev

poles.

s\

-0.443,

s' 2S

-0.222 /0.866

Finally, the denormalized Chebyshev poles are obtained by multiplying

the normalized ones by cosh (ik
1.1155 so that the denormalized poles
are sx
-0.494 and $2S
-0.249 /0.972.

H(s)

is

H(s)

then

^02
(s

+ 0.494)(s + 0.249 - ;0.972)(s + 0.249 +

(13.52)'
v

0.502

=
s*

0.992s*

/0.972)

1.255s

0.502

In Fig. 13.8, the amplitude responses of the Chebyshev and an n

Butterworth filter are shown.

Monotonic

filters

with

optimum

cutoff
with Chebyshev niters, the following
can be said. The Butterworth response is a maximally flat, monotonic
response, whereas the Chebyshev response is equal ripple in the pass band.
In the stop band, the Chebyshev response falls off more rapidly than the
Butterworth (except when e is very, very small). In this respect, the Chebyshev filter is a better filter than the Butterworth. However, as we shall see
in Section 13.5, the transient response of the Chebyshev filter is very poor.
If we require sharp cutoff characteristics for a given degree n, however,
the Butterworth filter is quite unsatisfactory. In 1958, Papoulis 8 proposed

In comparing Butterworth

A. Papoulis, "Optimum

March

1958, pp. 606-409.

filters

Filters with

Monotonic Response," Proc. IRE, 46, No.

3,

380

Network

analysis

and synthesis
Radian frequency,

02

0.1

0.6

0.4

0.3

o>

0.8

iI

'-;>< \

-1

2
\

-\\

\\

Amplitude response of n

-/

-8

Chebyshev

1.0-db ripple

in

with

filter

pass band

Buttterworth respons>e (n

\
3)
\\

\\

-10

\\

11
\\
I

12

13

-14
FIG.

13.8.

= 3 Chebyshev filter with 1.0-decibel ripple in pass

= 3).

Amplitude response of n

band and Butterworth response

(n

a class of filters called Optimum or

"L"

filters,

which have the following

properties:
1.

2.

The amplitude response is monotonic.

The fall-off rate at m cutoff is the greatest possible,

if monotonicity is

assumed.
3.

The

zeros of the system function of the

L filter

Recall that the magnitude response of a low-pass

infinity can be expressed as

M(o>)

filter

=
[1

with

all

zeros at

(13.53)

+JV)l*

Let us denote the polynomial generating the

/(>*)

are all at infinity.

L(ft> 8)

L filter by
(13.54)

Topics

in filter

design

381

Ln(mx) has the following properties:

The polynomial

=
L(l) =

LJO)

(a)

(b)

^^0

(0

dco

dL

>

2
>

=M

(M maximum)

dco

and c are the same as for the Butterworth generating polynomial/^*) = to2 ". Property c insures that the response M(a>) is monotonic and property d requires that the slope of L n(caP) at tu = 1 be the
Properties a, b

steepest to insure sharpest cutoff.

Papoulis originally derived the generating equation for the polynomials

Ln (for n

odd) to be

Ln((ot)=

where n = 2k
4
first kind

dx

a<p<(a;)

Llo

(13,55)

and the PJx) are the Legendre polynomials of the

=1
P^x) = x
P2(x) = K3* - 1)
P9(x) = i(5x* - 3x)
/>,>(*)

and the constants

a

a, are given

by

===
3

(1356)

2k

= 1=
1

V2(fc

(13.57)
1)

Later Papoulis 6 and, independently, Fukada,' showed that the evenordered L n polynomials can be given by

rim*-!

(a>8)

(*

-J

2k

1)[2>

-|

*>(*)]

dx

(13.58)

E. Jahnke and F. Emde, Tables of Functions, Dover Publications, New York, 1945.
A. Papoulis, "On Monotonic Response Filters," Proc. IRE, 47, February 1959,

332-333.

M. Fukada, "Optimum

Trans.

IRE, CT-6, No.

3,

Filters of Even Orders with Monotonic Response,"

September 1959, 277-281.

382

Network

anal/sis and synthesis

where the constants at are given by:

Case 1 (k even)
a

a*

(13.59)

V(fc

l)(fc

2)

V(fc

l)(k

2)

polynomials up to n

7 together with

2fc

1
1

M=

Case 2 (k odd):
a*
3
flo

+l
==.*
= 2
=
7

2fc

Fukada tabulated the

dLjai^ldco evaluated at
the cutoff. This

To

is

LB(w 2)
to

shown

(13.60)

to give an indication of the steepness of

in Table 13.4.

L filter, we must

obtain the system function H(s) for the

factor the

equation for his2) and choose the Hurwitz factors as H(s).

(13.61)

For example, for n

3,

the magnitude response squared

M\a>)

is

=
1

L^co*)
(13.62)
1

Substituting

a>*

<w = s we obtain
2
=
) = H(s) H(-s)
2

3>

3eo

fc(s

TABLE
L(a> 2

s*

3s

(13.63)
35*

13.4

Polynomials

dLn(l)
dm

Ln(a>2)

CO*

- 3o>4 + <u2
4
6eo 8 - 6o>* + 3to
10 - 40a>8
20o
+ 28a>8 - 8a>4 + to2
12 - 120a> 10 + 105e8 - 40w* + 6o>4
50to
10 - 355o>8 + 105a>
175*o M - 525a> + 615>

3o>8

12
18

24

-15eo4

a)

32

Topics

in filter

design

383

Frequency, rad/sec

02

0.1

0.3

0.4

0.6

0.8

-2

-4

1.0

Amplitude i esponsecrf
OpB mum vers
Butbsrworth fit ers

-6

\\

\\

-8

\\

\\

tf-10
0.4

|-12

-14
-16

0.5

Frequency
0.6
0.8

1.0

1.2

\ \

\\
y

-1.0

\\
\\

-18
\

-20

ft-

\
A

-22
Explanded
scale

-24

\
\

\\

-26
FIG.

After

we

13.9.

Amplitude response of Optimum versus Butterworth

factor A(s*),

we

if(s)

filters

obtain
0.577

=
s*

1.31s

1.359s

+ 0.577

(13.64)

where the numerator factor, 0.577, is chosen to let the d-c gain be unity.
poles of H(s) are sx = -0.62; s2>3 = -0.345 ;0.901. The amplitude response of third-order Optimum (L) and Butterworth niters are
compared in Fig. 13.9 Note that the amplitude response of the Optimum
filter is not maximally flat, although still monotonic. However,
the cutoff
characteristic of the Optimum filter is sharper than the cutoff of the

The

Butterworth

filter.

Linear phase filters

Suppose a system function

is

given by

H(s)

where

= Ke~'T

K is a positive real constant.

Then

(13.65)

the frequency response of the

system can be expressed as

H{jm)

Ke-"" T

(13.66)

Network

384

analysis and synthesis

so that the amplitude response M(<co) is a constant K,

sponse

#w)
is linear

in

= -mT

{e(t),

re-

(13.67)

The response of such a system

co.

the transform pair

and the phase

to

an excitation denoted by

E(s)} is

R(s)

so that the inverse transform

= K E(s)e-' T

r(t)

(13.68)

can be written as

KO-c-W)]
= Ke(t -

T)u(t

(1369)

T)

We see that the response r(t) is simply the excitation delayed by a time
and multiplied by a constant. Thus no

signal distortion results

T,

from

transmission through a system described by H(s) in Eq. 13.65. We note

further that the delay T can be obtained by differentiating the phase

response

<f>((o),

by

<u;

that

is,

Delay

_ - f^2> - r

(13.70)

dco

Consequently, in a system with linear phase, the delay of the system is

obtained by differentiating the phase response #co).
system with linear phase and constant amplitude is obviously desirable
from a pulse transmission viewpoint. However, the system function H(s)

in Eq. 13.65

a delay

only realizable in terms of a lossless transmission line called

If we require that the transmission network be made up of

is

line.

lumped elements, then we must approximate H(s)

function in s. The approximation method we
due to Thomson. 7

= Ke~ ,T by

a rational

shall describe here is

We can write H(s) as

(13.71)

o
sinh

sT + cosh sT

Let the delay T be normalized to

where K is chosen such that H(0)
denominator of H(s) by sinh s
and
unity and let us divide both numerator

=1.

to obtain

Kjsmhs
coth
'

W.

E.

Thomson, "Network with Maximally

1952, 256-263.

(13?2)

Flat Delay," Wireless Engrg., 29, Oct.

If sinh s

and cosh * are expanded

coshs

in

power

Topics

in filter

series,

we have

s*
s*
s*
= l++-+-+

From these series expansions, we

385

---

(1373)

,'

.'
6

s
s
sinh5- S + - + - + - +
s

design

'--

then obtain a continued fraction expan-

sion of coth s as

coth

=-+
3

(13.74)

~*

z+...
5

If the continued fraction is terminated in

n terms, then H(s) can be

written as

where

Bn(s) are f?es.>/ polynomials defined by the formulas

2*0=1

Bt = s+1

From these

formulas,

(13.76)

2?

- (2/i - 1)3^ + s*Bn_t

we

obtain

= s* + 3s + 3
2?, = j + 6s* + 15s +
if,

(13.77)

15

Higher order Bessel polynomials are given in Table 13.5, and the roots of
Bessel polynomials are given in Table 13.6. Note that the roots are all in
the left-half plane. A more extensive table of roots of Bessel polynomi8
als is given by Orchard.
The amplitude and phase response of a system function employing an
unnormalized third-order Bessel polynomial

W=

Ms)

s*

^+

6s*

15s

(13.78)

15

H. J. Orchard, "The Roots of Maximally Flat Delay Polynomials"

on Circuit Theory, CT-12 No. 3, September 1965, 452-454.

IEEE Trans,

Network

386

analysis

and synthesis

TABLE

13.5

Coefficients of Bessel Polynomials

bo

bt

b*

b7

1
1

15

15

105

105

10

945

945

6
45
420

105

15

10,395

10,395

4,725

1,260

210

21

135,135

135,135

62,370

17,325

3,150

378

28

are given by the solid lines in Figs. 13.10 and 13.11. These are compared
with the amplitude and phase of an unnormalized third-order Butterworth
function given by the dotted lines. Note that the phase response of the
constant-delay function is more linear than the phase of the Butterworth
function. Also, the amplitude cutoff of the constant-delay curve is more

gradual than that of the Butterworth.

TABLE

13.6

Roots of Bessel Polynomials

Roots of Bessel Polynomials

1

2
3

-1.0 +y0
-1.5 y0.866025
[-2.32219 +y0
1.83891 yl.75438
i
/ -2.89621 y0.86723
\ -2.10379 y2.65742
-3.64674 +j0
-3.35196 yl.74266
[-2.32467 /3.57102
C

-4.24836 y0.86751
-3.73571 /2.62627
-2.51593 y4.49267

-4.97179 +j0
-4.75829 yl .73929
-4.07014 y3.51717
1-2.68568 y5.42069
r

Topics

0.1

0.4

0.2

in filter

Frequency, rad/sec
0.6
0.8 1

-2

387

design

6 7

\
>^

-4

-Bessel

-6

-8

,-10
B it* srv fO rth

h
!

\
\

-14

-16

-18

<

-20

-22

-24

\
\

-26
FIG. 13.10. Amplitude response of n

= 3 Bessel and Butterworth filters.

Frequency, rad/sec
0.2

0.4

FIG.

0.6

13.1 1.

0.8

1.0

1.2

Phase responses of low-pass

1.4

filters.

1.6

1.8

2.0

388

13.5

Network

analysis

and synthesis

TRANSIENT RESPONSE OF LOW-PASS FILTERS

we

compare the

transient response of the filters

discussed in Section 13.4. In particular,

we will compare the step response

In this section

will

of the niters according to the following figures of merit:

1.

Rise time

The

rise

time of the step response

the time required for the step response to rise

is

defined here as

from 10%

to

90%

of

its

final value as depicted in Fig. 13.12.

Ringing is an oscillatory transient occurring in the response

of a filter as a result of a sudden change in input (such as a step).
quantitative measure of the ringing in a step response is given by its
2. Ringing.

settling time.
3. Settling time.

The

settling

step response does not differ

time

is

from the

that time

final

t,

value by

beyond which the

more than 2%,

as depicted in Fig. 13.12.

Delay time is the time which the step response

of its final value as shown in Fig. 13.12.
5. Overshoot.
The overshoot of the step response is defined as the
difference between the peak value and the final value of the step response
4.

Delay time,

requires to reach

50%

(see Fig. 13.12) expressed as

a percentage of the

FIC.
tB

12

10

Time

t,

sec

13.12. Figures of merit for step response.

delay time.

final value.

ta

= rise

time;

/.

setting time;

Topics

in filter

389

design

Most of the foregoing figures of merit are related to frequency response,

particularly bandwidth and phase linearity. Some of the quantities, such
related to each other but have
Let us examine qualitatively the
relationships between the transient response criteria just cited and

and delay time are intimately

as rise time

rather tenuous ties with overshoot.

frequency response.
Rise time and bandwidth have an inverse relationship in a filter. The
wider the bandwidth, the smaller the rise time; the narrower the bandwidth, the longer the rise time. Physically, the inverse relationship could

be explained by noting that the limited performance of the filter at high

down the abrupt rise in voltage of the step and prolongs
the rise time. Thus we have

frequencies slows

TB X

BW = Constant

(13.79)

Rise time is a particularly important criterion in pulse transmission. In

an article on data transmission,* it was shown that in transmitting a pulse

of width 7\ through a system with adjustable bandwidths, the following

results were obtained:
Rise Time

Bandwidth
(/.

- UTJ

(milliseconds)

0.5

f.

The

table

shows a

2/

0.25

3/c

0.16

4/

0.12

y.

0.10

definite inverse relationship

between

rise

time and

bandwidth.

definition of time delay

is

given by Elmore as the

first

moment

or

centroid of the impulse response

Jo
-r

h{t) dt

(13.80)

provided the step response has little or no overshoot. Elmore's definitio

of rise time is given as the second moment

TB

= \2nj\t - TDf h(t) a]

(13.81)

R. T. James, "Data TransmissionThe Art of Moving Information,"

Spectrum, January 1965, 65-83.

IEEE

Network

390

anal/sis and synthesis

These definitions are useful because we

can obtain rise time and delay time directly

from the coefficients of the system function

H(s). Without going into the proof, which
is in Elmore and Sands, 10 if H(s) is given as

His)
FIG.

13.13.

R-C network.

+ lS + a *s2 +
+ b s + 6js +
x

'

+ a*s n
+ b ns m
(13.82)

the time delay

and the

rise

time

bl

ax

af

2(a 2

TD =

TD i%
TR =

For the R-C network

{2tt[V

in Fig. 13.13, H(s)

H(s)

- b )]}*

(13.84)

is

V(s)
I(s)

(13.85)

sRC

TD = RC_
TB = yJlnRC

so that

It

(13.83)

is

(13.86)

should be emphasized that Elmore's definitions are restricted to step

responses without overshoot because of the moment definition. The more

general definition of rise time is the 10-90% one cited earlier, which has

no formal mathematical
Overshoot
excess gain

is

definition.

generally caused

by "excess" gain

we normally mean a magnitude

at high frequencies.

characteristic with a

By

peak

such as the shunt peaked response shown by the dashed curve in Fig.
13.14. A magnitude characteristic with no overshoot is the magnitude
characteristic of

an R-C interstage shown by the

solid curve in Fig. 13.14.

Log frequency

FIG.
10

V,

13.14.

Comparison of shunt-peaked and simple R-C magnitudes.

W. C. Elmore and M. Sands, Electronics, National Nuclear Energy

McGraw-Hill Book Company, New York, 1949, pp. 137-138.

1,

Series,

Div.

Topics

in filter design

391

1.2
1.1

1.0

0.9

?0.8
N

=5 0.7

5
c

0.6
0.5

In

f 6.4
I

J 3 ~A> = 7/il=lo"

0.3
0.2
0.1

10

Time

FIG.

13.15.

t,

12

16

14

18

20

sec

Step response of normalized Butterworth low-pass

filters.

The step responses of the n = 3, n = 7, and n = 10 Butterworth filters

are shown in Fig. 13.15. Note that as n increases, the overshoot increases.
This

because the higher order Butterworth niters have flatter magnitude

(i.e., there is more gain at frequencies just below the cutoff).
Ringing is due to sharp cutoff in the filter magnitude response, and is
accentuated by a rising gain characteristic preceding the discontinuity.
is

characteristics

an n = 3 Bessel (linear phase) filter is compared

an n = 3 Chebyshev filter with 1-decibel ripple in
Fig. 13.16. We cannot compare their rise times since the bandwidths of
the two filters have not been adjusted to be equal. However, we can
compare their ringing and settling times. The Chebyshev filter has a
sharper cutoff, and therefore has more ringing and longer settling time
than the Bessel filter. Note also the negligible overshoot of the Bessel

The

step response of

to the response of

filter

that

The

is characteristic

of the entire class of Bessel

decision as to which

filter is

best depends

filters.

upon

the particular

In certain applications, such as for transmission of music,

not important. In these cases, the sharpness of cutoff may be the

situation.

phase

is

dominant factor so that the Chebyshev or the Optimum filter is better

than the others. Suppose we were dealing with a pulse transmission
system with the requirement that the output sequence have approximately
the same shape as the input sequence, except for a time delay of T
Tt Tx , as shown in Fig. 13.17a. It is clear that a filter with a long rise

time

is

not suitable, because the pulses would "smear" over each other

Network

392
1.1

1.0

/
t
t

0.9

0.8

and synthesis

analysis

'

^ ^ s

!
1

0.7

S 0.6

3 Bessel fitter
-Step response of n
-Step response of n 3 Chebyshev
filter with 1-db ripple

nr.
a 0.5
i

<0.4

1
|

0.3

0.2

~r

0.1

ly

12

10

14

16

18

20

Time, sec

FIG. 13.16. Comparison of filter transient responses.

System
with short
rise

and

settling

Ti

times
Input pulse train

ALA
T2

Output pulse

train

(a)

System
with long
rise

and

settling

Ti

times

Output pulse train

Input pulse train

(b)

FIG. 13.17. Smearing of pulses in systems with long

rise

and

setting times.

as seen in Fig. 13.176. The same can be said for long settling times.
Since a pulse transmission system must have linear phase to insure undistorted harmonic reconstruction at the receiver, the best filter for the

system
13.6

is

a linear phase

filter

with small

rise

and

settling times.

A METHOD TO REDUCE OVERSHOOT

IN FILTERS

method to reduce the overshoot and ringing of a

filter
filter step response. The step response of a tenth-order Butterworth
We
is
about
overshoot
18%.
the
seen
that
is shown in Fig. 13.18. It is

We

present here a

Topics

393

in filter design

\2

1.0

0.8

tep re sponsc

0.6

0.4

/
/

02

.'

Second

derivative

"^ ^>

r
/

\
\

-1.2

10
12
Time t, sec

16

14

18

20

FIG. 13.18

note that after the

first

peak, the ringing of the step response has an

approximate sinusoidal waveshape. Let us now consider the second

derivative of the step response shown by the dashed curve in Fig. 13.18.
Beyond the first peak of the step response, the second derivative is also
(approximately) sinusoidal, and

is

negative

when

when

the step response

is

than unity.
If we add the second derivative to the step response, we reduce the overshoot and ringing. 11
Suppose a(0 is the step response and H(s) is the system function of the
greater than unity,

filter.

The

and

positive

the step response

is less

corrected step response can be written as

a1(0

= (0 +

L>

(13.87)

where K is a real, positive constant. Taking the Laplace transform of

Eq. 13.87, we have

tW*)]

Xs

L
(13.88)

H(s)
s
11

F. F.

Kuo, "A Method to Reduce Overshoot

No. 4, December 1962, 413-414.

Theory, CI-9,

in Filters," Trans.

IRE on

Circuit

Network

394

analysis and synthesis

1.2

/%

1.0

'

--.

^-.^

k
i.

0.8

lo.6

<E

Tenth order Butterworth

Zeros at
Zeros at

0.4

= i

1.0

a>= *

1.5

a)

Jtt

0.2

y/f
Jl

12
10
Time t, sec

16

14

18

20

FIG. 13.19

From
s

Eq. 13.88

we

see that

by adding a pair of zeros on the jm

axis at

//v K, the overshoot and ringing are reduced.

l/\K must in general be greater than

For normalized Butterworth filters the
bandwidth is co = 1 so that K <, 1. The factor AT also controls the amount
of overshoot reduction. If K is too small, adding the zeros on they'co axis
For low-pass

niters, the factor

the bandwidth of the system.

-o
-2
-4

-6
-8
-10
.o
-o

-12

c -14
<3

_i 6
-18

-20

"

Tenth order Butterworth

Zeros at co = t 1.0
Zeros at w = * 1.5
Zeros at co = 1.8: with
original bandwidth also
equal to 1.8

-22

-24

-26
-28
0.1

0.2

0.3

0.4

0.6

0.8 1.0

Radian frequency,

FIG. 13.20

1.5

2.0

3.0

Topics

in filter

395

design

will have negligible effect. Therefore the zeros should be added somewhere near the band edge. Figure 13.19 shows the effects of adding
zeros at <w = 1 (i.e., right at the band edge), and at m = 1.5. We see
that the further away the zeros are placed from the band edge, the less

they will have. The addition of the zeros will decrease the 3-decibel
bandwidth of the filter, however, as seen in Fig. 13.20. Therefore, a
effect

compromise must be reached between reduction

in

bandwidth and

reduction in overshoot.

An effective way to overcome this difficulty is to scale up the bandwidth

by a factor of, for example, 1.8. Then the zeros are placed at co = 1.8,
which will reduce the 3-decibel bandwidth to approximately its original
figure <w = 1.0, as shown in Fig. 13.20. The overshoot, however, will be
reduced as though the zeros were at the band edge.

A MAXIMALLY FLAT DELAY AND CONTROLLABLE

MAGNITUDE APPROXIMATION

13.7

we

examine an interesting result, which is due to

phase approximation with controllable magnitude. In Section 13.4 we discussed approximation of a
flat delay using Bessel polynomials. The resulting rational approximant
was an all-pole function (all transmission zeros at s = oo), whose denominator was a Bessel polynomial. There was no control of the magnitude using the all-pole approximant. In Budak' s method the magnitude
is controllable, while the phase is as linear as the standard Bessel approxiIn this section

Budak. 1 * The

will

result deals with linear

mation.

Budak's approximation is obtained by introducing the parameter k to

two parts such that

split e~~* into

*-

= pSr.

<

<. 1

(13.89)

and then approximate independently c-** and e~lk~lU with all-pole Bessel
polynomial approximations. Thus the resulting approximation for e~*
will have Bessel polynomials for both numerator and denominator. The
poles of the e~ * -1) ' approximant will be the zeros in the final approximant,
(

while the poles of the e~** approximant remain as poles in the final

approximant. For realizability, the degree of the c- <t-1) * approximant

should be less than the degree of the e ** approximant.

11

A. Budak,

Trans,

"A Maximally Flat Phase and Controllable Magnitude Approximation,"

of IEEE on

Circuit Theory,

CT-12, No.

2,

June 1965, 279.

Network

396

anal/sis

and synthesis

A0

M(o>)
1.0

1.0

-0.6

T*
0.8

0.8

0.7^

-1.0

ft

0.6

0.6

(Bessel)"^

A=1.0_
0.4

0.4

k-

0,8-

02

km 0.7-j^
*-0.6tC

0.2

(a)

(b)

FIG.

As an example,

13.21

consider the approximation with three zeros and four

poles.

105
(ks)*

10(fcs)

+ 45(fcs)* +

[(fe

(13.90)

105

l)sf

6[(fc

l)s]

15[(ik

l)s]

(13.91)

15

then perform the operations as indicated by Eq. 13.89 to obtain

-._ 7{[(fc -

15

g-tt-i).^

We

105(jfcs)

(ks)*

1)5]'

- l)s] + 15[(fe - l)s] + 15}

+ 45(fcs) + 105(Jb) + 105

6[(k

10(fcs)

In Fig. 13.21a the magnitude characteristic of Eq. 13.92 is plotted with

The phase characteristic is given in terms of deviation

as a parameter.

of phase from linearity A<f> = eo <(co) and is shown in Fig. 13.216.

The improvement in phase linearity over the all-pole Bessel approximation
k = 1 is shown by these curves. Note as the bandwidth is increased
(A:

decreasing), phase linearity is improved.

response of Eq. 13.92 also with

decreasing

A:

Figure 13.22 shows the step

as a parameter.

Since the effect of

increases bandwidth, the corresponding effect in the time

domain is to decrease rise time.

Budak also observes that as k decreases from unity, the poles and zeros
migrate to keep the phase linear. The zeros move inward from infinity
along radial

lines,

while the poles

move outward along

radial lines.

Topics

in filter

design

397

l.l

1.0

k> 0.6-

0.9

0.8

*>0.7-

-km

).8

0.7

0.6

_*-1.0
(Bessel)

0.5

%
0.4

0.3

02
0.1

0.0

-0.1

-0,2

02

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

FIG. 13.22

13.8

SYNTHESIS OF LOW-PASS FILTERS

Given the system function of the low-pass filter as derived by the

methods described in Section 13.4, we can proceed with the synthesis of
the filter network. If we consider the class of filters terminated in a 1-Q
load, and if we let the system function be a transfer impedance,
Z(s)

or a transfer admittance

= -^a-

Yn(s) =
1

(13.93)

(13.94)

we can synthesize the low-pass filter according to methods given in

Chapter 12. For example, consider the n = 3 Optimum (L) filter function

Network

398

analysis

and synthesis

given as a transfer impedance

0.577

Z21W

1.31s

1.359s

(13.95)

0.577

We see that the zeros of transmission are all at infinity.

ator of

Z21

is

we

even,

divide both numerator

odd part of the denominator

s3

Since the numerand denominator by the

Thus we have

1.359*.

0.577

Za
s
Z S2

The

structure of the low-pass

infinity is given in

Chapter

reactance structure. This

we

fraction expansion of l/ 22
1.31s 2

(13.96)

0.577

1.359s

with three zeros of transmission at

filter

12.

1.359s

1.31s

We

must synthesize

to give the

it

accomplish through the following continued

+ 0.577 )ss + 1.359s (0.763s

s8 + 0.440s
0.919s ) 1.31s8
1.31s

+ 0.577 ( 1.415s

0.577) 0.919s ( 1 .593s

0.919s

The optimum filter is shown in Fig.

For the n

13.23.

= 3 Butterworth filter

given by the transfer impedance

Z2l(s)

we have

z n (s)

2s

2s

(13.97)

+ 2s-rl
,.

=
s

We

~ss +

2s

-t 7T7s

(13.98)

then synthesize M (s) by a continued fraction expansion to give the

filter

shown

in Fig. 13.23.

(a)

Optimum

(b) Butterworth

filter

FIG. 13.23

filter

Topics in

filter

>ia v2

c7 :i

FIG. 13.24. Canonical form for

niters described in

Tables 13.7, 13.8, and 13.9.

In Tables 13.7, 13.8, 13.9 are listed element values (up to n

single-terminated Butterworth, Chebyshev (1-decibel ripple),
filters,

respectively. 13

transfer

impedance

desired,

we

vice versa.

399

design

7) for

and Bessel

These apply to the canonical realization for a

Z21(s)

shown

simply replace

all

The element values

in Fig. 13.24.

If a

shunt capacitors by
all

Y21(s)

realization is

series inductors

and

carry over.

we will consider some examples of synthesis of double

To stimulate the curiosity of the reader, note that the
transfer function V2 jV of the network in Fig. 13.25 is

In Chapter 14,
terminated

filters.

voltage-ratio

precisely the n

3 Butterworth function.

Recall that in Chapter 12,

when we cascaded two

networks, the overall system function

(s)

constant-resistance

was the product of the

in-

We

can apply this property to

networks which are not constant-resistance if we place an isolation amplifier between the networks, as shown in Fig. 13.26. Since pentodes provide
the necessary isolation, our task is simplified to the design of the individual
structures H^s),
2 (s),
n (s), which we call interstage networks.
dividual system functions

Some common

t (s)

2 (s).

shown in Fig. 13.27. In Fig.

a structure known as the shunt-peaked network is shown. The transfer impedance of the shunt-peaked network is
interstage structures are

1 3.27a

Za(s)

FIG. 13.25. n

= -1

13,

RjL

3 double-terminated Butterworth

u More extensive tables are given in L.

and Synthesis, Chapter

(13.99)

Cs 2 + sR/L+(llLC)

filter.

Weinberg's excellent book, Network Analysis

McGraw-Hill Book Company, 1962.

400

Network

analysis

and synthesis

TABLE

13.7

Normalized Element Values for a Single Terminated

Butterworth Filter

Cx

Lt

c,

L*

Lt

1.000

0.707

1.414

0.500

1.333

1.500

0.383

1.082

1.577

0.309

0.894

1.382

1.694

1.545

0.259

0.758

1.202

1.553

1.759

1.553

0.222

0.656

1.055

1.397

1.659

1.799

C7

1.531

TABLE

1.558

13.8

Normalized Element Values for a Single Terminated

Chebyshev Filter with l-db Ripple

Cx

L,

C,

L4

Cs

0.509

0.911

0.996

1.012

1.333

1.509

1.050

1.413

1.909

1.282

1.067

1.444

1.994

1.591

1.665

1.077

1.460

2.027

1.651

2.049

1.346

1.083

1.496

2.044

1.674

2.119

1.649

TABLE

1.712

13.9

Normalized Element Values for a Single Terminated

Bessel Filter

Lt

c,

Z*

c.

1.000

0.333

1.000

0.167

0.480

0.833

0.100

0.290

0.463

0.710

0.067

0.195

0.310

0.422

0.623

6
7

0.048

0.140
0.106

0.225

0.301

0.382

0.170

0.229

0.283

0.036

A.

0.560
0.349

0.511

Topics

in filter design

401

Pentode
Interstage

Interstage

Interstage

network

network

network

FIG.

Pentodes used as isolation amplifiers.

13.26.

We sec that Ztl(s) has a real zero and a pair of poles which may be complex depending

upon

the values of R, L,

R-C interstage is shown, whose

transfer

Zn(s) Observe that

all

the

filter

and C. In Fig. 13.27A, a simple

impedance is

Cs +

(13.100)

1/RC

transfer functions considered

up

to this point

made up of pairs of conjugate poles and simple poles on the a axis.

is clear that if we cascade shunt-peaked stages and R-C stages, we can

are
It

and C elements to give the desired response characteristic.

The only problem is to cancel the finite zero of the shunt-peaked stage.
For example, if we wish to design an amplifier with an n = 3 low-pass
Butterworth characteristic, we first break up the system function into
adjust the R, L,

complex pole pairs and

real pole terms, as given

Z(s)

by

+ s + l)(s + 1)
1
s+
1
s* + s+ls + ls +
(s*

(13.101)

We

then associate the individual factors with shunt-peaked or simple

stages and solve for the element values. The n *= 3 Butterworth
amplifier is given in Fig. 13.28.

R-C

ci

ci ^*
.R

(o>

FIG.

13.27. (a)

Shunt-peaked interstage.

(b)
(f>)

R-C interstage.

Network

402

analysis

and synthesis

FIG.

Butterworth amplifier.

13.28.

MAGNITUDE AND FREQUENCY NORMALIZATION

13.9

In Section 13.8,
cutoff frequency of

we

discussed the synthesis of low-pass niters with a

rad/sec and a load impedance of

1 Q. Filters designed
with these restrictions are considered to be normalized in both cutoff
frequency and impedance level. We will now discuss methods whereby

the normalized

which meet arbitrary

Let us denote by a
subscript n the normalized frequency variable s n and the normalized
element values L, jR, and C. The normalized frequency variable s n is
related to the actual frequency s by the relation
filters

cutoff frequency

can be converted into

and impedance

filters

level specifications.

(13.102)

where <u the normalizing constant, is dimensionless and is often taken to

be the actual cutoff frequency.
Since the impedance of an element remains invariant under frequency
normalization, we obtain the actual element values from the normalized
values by setting the impedances in the two cases equal to each other. For
example, for an inductor, we have
,

sL

From

this

equation

we then

= sL =

a>o5 n

(13.103)

obtain the denormalized value of inductance

as

L-^

(13.104)

from the impedance l/sn C of a frequency normalized capacitor

we obtain the denormalized value of capacitance through the equation

Similarly,

C,

1
sn

Cn

(13.105)

sC

"

Topics

so that the actual value of the capacitance

C=

in filter

design

403

is

(13.106)

>o

Since resistances, ideally, are independent of frequency, they are unaffected by frequency normalization.

Consider,

impedance
impedance

next,

level

Z is

impedance denormalization.

Z
where R

Rn

is

Suppose the actual

should be Rq ohms instead of 1 Q. Then a denormalized

related to a normalized impedance Z n by
2?oZ

(13.107)

taken to be dimensionless here. Thus, for a normalized resistor

the denormalized (actual) resistance

R=

RoR

For an inductance, the corresponding

sL

is

(13.108)

relationship

= R^sL n)

so that the actual inductance value

(13.109)

is

L = R Ln
Similarly, for a capacitor

is

(13.110)

we have

so that the actual capacitance

_L

= Jk

sC

sC

(13.111)

is

C=

(13.112)

For combined frequency and magnitude denormalization, we simply

combine the two sets of equations to give

R = RoR n
C=

R^o

(13.113)

L = ^O^n
co a

Let us consider an actual example in design. In Section 13.8, we

Zn with an n 3 Butterworth amplitude
characteristic with a cutoff frequency of 1 rad/sec and a load impedance of
1 ii. Let us redesign this filter for a cutoff frequency of 10* rad/sec to work
synthesized a transfer impedance

Network

404

and synthesis

anal/sis

mh

66.7

Q>

^=0.3 nf

=4=0.1 nf

500 a

FIG. 13.29. Denormalized low-pass

into a load of 500 Q.

From the

original

V2

filter.

network in Fig. 13.23, we take the

element values and denormalize with the normalizing factors,

and Rt

Then

<o

10*

500.

the denormalized element values are

R=

C,=

- 500

500*
i

500004)

0.1 /*f

(13.114)

L = 1(500) = 0.0067 h
10,000

C,

ST
4

500(10

The

final design is

13.10

shown

0.3^f

in Fig. 13.29.

FREQUENCY TRANSFORMATIONS

Up to this point, we have discussed only the design of low-pass filters,

while neglecting the equally important designs of high-pass, band-pass,

We

not by
as a
introducing new design procedures but through a technique
low-pass
normalized
frequency transformation, whereby, beginning from a
Using frequency transfilter, we can generate any other form of filter.

and band-elimination

filters.

will

remedy

this situation here,

known

formations, the elements of the normalized low-pass filter are changed

into elements of a high-pass, band-pass, or band-elimination filter.
Analytically,

a frequency transformation simply changes one L-C

I^C driving-point function. Therefore,

driving-point function into another

the transformation equations

must be LrC functions themselves. Also,

low-pass filters, the transformation

equations include built-in frequency denormalization factors so that the
the
resulting networks need only be scaled for impedance level. Consider

since

we proceed from normalized

Topics

in filter design

405

simplest transformation equation, that of low-pass to high-pass, which

= a>

is

(13.115)

where sn represents the normalized low-pass frequency variable, s is the

regular frequency variable, and to is the cutoff frequency of the high-pass
filter. In terms of real and imaginary parts, we have

+ jm =

>o
ff +./'>

(13.116)

we are interested principally

we let <r = so that

Since
axis,

in

how

the j<o n axis

maps

into theycu

co

(13.117)
a>.

which is the equation that transforms normalized low-pass filters to

denormalized high-pass filters. From Eq. 13.117 we see that the point
wn = 1 corresponds to the point m = a> It is also clear that the
transformation maps the segment \a>\ ^ 1 on to the segments denned by
.

mo ^
^ > as shown in Fig. 13.30.
Now let us see how the frequency transformations change the network
For convenience, let us denote the normalized low-pass network
elements with a subscript n, the high-pass elements with a subscript h, the
band-pass elements with a subscript b, and the band-elimination elements
with a subscript e. For the low-pass to high-pass case, let us first consider
the changes for the capacitor C. The transformation is given by the
elements.

ju

JVn

+j

plane

/<*>

* plane

<r*

-J

FIG.

13.30.

-Jox>

Low-pass to high-pass transformation.

Network

406

anal/sis and synthesis

equation

= LhS

Cs n

For the inductor

L n we
,

fo

(13.118)

C,

have
(13.119)

Chs

We

observe that a capacitor changes into an inductor and an inductor

changes into a capacitor in a low-pass to high-pass transformation (Fig.
13.3 1).

The element values of the high-pass filter

normalized low-pass

filter

Lh

(13.120)
<o

Cn

(o

Ln

Ch =

and

(13.121)

From

Consider the following example.

Butterworth

filter

with

pass

filter

level

of 500 Q.

given in Fig. 13.23,

cutoff frequency

its

From

the low-pass

High-pass

Low-pass

are given in terms of the

elements as

cu

let

the normalized third-order

us design a corresponding high-

filter,

and the impedance

we can draw by inspection the
10 rad/sec

Band-pass

Band-elimination

1*1

o-^fiflp-o

1{

Ca

1_
<>

Cbl

npnr- -H(BW
BW
2L
<">0

Cd = LnBW

r. BW
Lh woC
n
o-'lHRP-o
-

La
o-j^nnp-

Cc2

CnBW
Cb2=

FIG.

13.31.

-gyp

Element changes resulting from frequency transformations.

Topics

Ch =

1.5

x 10~ 9

in filter

407

design

Hf-

FIG.

13.32. Transformation of low-pass filter in Fig. 13.23 into high-pass

high-pass-filter circuit

shown

in Fig. 13.32.

Its

filter,

element values are:

10 (i)
1

(13.122)

son
4

10 (f)
Next, let us examine the low-pass to band-pass transformation (also
an L-C function):
S

where,

if

+ -)
= BW
S(\a>
s I

<

13123 >

m ct and wp, denote the upper and lower cutoff frequencies of the
filter, BWis the bandwidth

band-pass

BW^cocz-cOd
and

<o

is

the geometric

mean of <o C2 and

m =

co cl

Vo)c2>ci

The low-pass

to band-pass transformation

the segments

m^ ^

|a|

^ eu^,

(13.124)

shown

maps

(13.125)

the segment

in Fig. 13.33.

\a> n \

<,

to

The normalized

low-pass elements are then modified according to the following equations

**.

bs + BWs
BW

(13.126)

408

Network

anal/sis and synthesis

\jo>

0
C1

+j
Plane

s plane

<Tn

-J
-<oo

FIG.

13.33.

Low-pass to band-pass transformation.

We

note that the inductor L is transformed into a series-tuned tank,

shown in Fig. 13.31, whose elements are given as

LM

BW
BW
CM =
>0

The capacitor C is transformed

whose elements are

(13.127)

Ln

into a parallel-tuned tank (Fig. 13.31),

LfiS

BW
(13.128)

C/h9

Let us transform the third-order Butterworth low-pass

BW =6x10*
rad/sec.

We

draw the band-pass

(T\ isg

13.34.

in Fig. 13.23

filter

Li

FIG.

filter

with a 1-ii impedance level, whose bandwidth is

4 x 10*
rad/sec, and its band-pass is "centered" at w

into a band-pass

Band-pass

filter

shown

filter

in Fig. 13.34

by the

rules

Ct

nroip

1(-

d=c3

"=r-Ci

yCi

Sjij

i*

transformed from low-pass

filter

v2

in Fig. 13.23.

Topics
given above.

The element

in filter

values of the band-pass

filter

design

409

are given in the

following equations:

10
= 0.75
U 6x
x io*m)
*

**

x KT'h

(4

Ci-

X
X

10*

10-*f

12

-^ = ^xl(T
6

10*

9
(13.129)

X 10At)

(4

10*

6X104

=
(4

10-*f

32

= o.25

x l(T*h

X lO^f)

Q = 6x

= 0.25

X KT*f

10*

through the transfor-

Finally, the band-elimination filter is obtained

mation
s.

BW
Is
\O)

BW and

(13.130)
eu

S I

manner similar to that for the band-pass

segment of the _/>,, axis in Fig. 13.35a
the
maps
filter. The transformation
in Fig. 13.356. For the low-pass
axis
ja>
on
the
shown
onto the segments

where

a>

are defined in a

je>

Jf>n

wo

-J

-o
-Wei

(a)

(b)

FIG. 13.35. Low-pass to band-elimination transformation.

410

Network

analysis

and synthesis

to band-elimination transformation we, therefore, have the following

element changes:

Ls
n
"

(s/L n BW)

ML BWs)
n

Ctls +

(1/L a s)

(13.131)

Cn s n

C nBW\a>

= Lsis

-\

C<gS

Observe that the normalized low-pass inductor goes into a paralleltuned circuit and the capacitor C goes into a series-tuned circuit, as
shown in Fig. 13.31. In Table 13.10, we have a composite summary of the
various transformations.

TABLE

13.10

Table of Various Frequency Transformations

Transformation
Equation

Low-Pass to

High-pass

Band-pass

Band-elimination

sn

BW

fe + ?)
Problems
13.1
figure.

Z =

VJIX for the filter shown in the

n
in order for \Z21 (ja>)\ to be maximally flat?

Find the transfer impedance

What should L be

PROB.

13.1

Topics
13.2 Find the poles of system functions with n
Butterworth characteristics. (Do not use the tables.)

Show

13.3

response

is

at

that the half-power point of a

m =

cosh

for e

/?*

in filter

3,

design

= 4,

41

and n

Chebyshev low-pass amplitude

1.

Determine the system function for the following

Ripple of \ db in band |o>| <> 1
(b) at co = 3, amplitude is down 30 db.
13.4

filter specifications:

(a)

Compare the slopes

13.5
(a)

/(a>*)

(6)

/(a,*)

(c)

/(>)

at

to

of the following polynomials (for n

= a>*
- iC(2a* = Ln(8).

Determine the polynomials

13.6

1)

+J =

L4(co 2) and

3):

Q(.)

!(>*).

13.7 Expand cosh j and sinh s into power series and find the first four terms
of the continued fraction expansion of cosh */sinh s. Truncate the expansion at
= 4 and show that H(s)
KjB^s).
13.8 Synthesize the n
ated in a l-O load.

Synthesize the low-pass

13.9
will

=
= 3 linear phase filter as a transfer impedance terminfilter,

which,

when terminated

have a transfer admittance whose poles are shown in the

PROB.

in

1-il resistor,

figure.

13.9.

Determine the asymptotic rate of falloff in the stop band of: (a)
filters; (b) linear phase filters.
13.11 Synthesize the n = 3 and n = 4 Butterworth responses as transfer
impedances terminated in a load of 600 SI with a cutoff frequency of 10* rad/sec.
13.12 Synthesize a Chebyshev low-pass filter to meet the following speci13.10

optimum

fications:

RL =

600

(a)

load

(b)

&-db ripple within pass band

resistor,

(c) cutoff

frequency

(d) at 1.5

x 105

rad/sec

10* rad/sec, the magnitude

must be down 30 db.

Network

412
13.13

analysis

Synthesize n

and synthesis

= 3 Optimum and linear phase filters to meet the following

specifications:
(a) load resistor
(b) cutoff

13.14

10*

frequency

Design an n

Q
10* rad/sec.

=4

Butterworth amplifier with the following

specifi-

cations:
(a)

impedance

(ft)

cutoff frequency

13.15

level

= 500 ft
= 10* rad/sec.

Synthesize a high-pass

filter

in a 10*-O load,

whose amplitude

frequency of

a>

for a given transfer admittance terminated

Optimum (L), with a cutoff

characteristic is

10* rad/sec.

Synthesize: (a) a band-pass filter; (b) a band-elimination filter, with

8 x 10* and o>ci
maximally flat (n
4) amplitude response with ce

13.16

2 x

10*.

w-

chapter 14

The

14.1

INCIDENT

AND

scattering matrix

REFLECTED

POWER FLOW

we will devote our attention to certain power relationand two-port networks. The characterization of a network in
terms of power instead of the conventional voltage-current description is a
helpful analytical tool used by transmission engineers. It is especially
important in microwave transmission problems where circuits can no
longer be given in terms of lumped R, L, and C elements. In the powerflow description, we are concerned with the power into the network, which
we call the incident power and the power reflected back from the load,
which is the reflected power. A convenient description of the network in
terms of incident and reflected power is given by the scattering matrix,
which is the main topic of discussion in this chapter.
It is convenient to think of incident and reflected power when dealing
with transmission lines. Therefore, we will briefly review some concepts in
transmission line theory. For a more comprehensive treatment of transmission lines, the reader is referred to any standard text on wave propagation. 1 Consider the transmission line shown in Fig. 14.1. The voltage
at any point down the line is a function of x, the distance from the source.
The parameters which describe the transmission line are given in the
In this chapter,

ships in one-

following:

R=
G=
L=

C=
1

resistance per unit length

conductance per unit length
inductance per unit length
capacitance per unit length

See, for example, E. C. Jordan, Electromagnetic

Prentice-Hall,

EngUwood Cliffs, New

Jersey, 1950.

413

Waves and Radiating Systems,

Network

414

anal/sis

and synthesis
l-

h:

FIG.

14.1.

Transmission

line.

Given these parameters, we can now define the impedance per unit
length as

Z = R+joL

(14.1)

and the admittance per unit length as

Y=G+jcoC
The

characteristic

impedance

of the line

is

(14.2)

given in terms of

z = Vz/r
and the propagation constant

Z and

Fas

(14.3)

is

sfZY

(14.4)

equations for
these definitions in mind, let us turn to the general
line
the
x
down
point
any
the current and voltage at

With

. Vfi->* + Vsyx
I(x) = !*" - iy

V{x)

(14.5)

the
/ refer to the incident wave at point * and
14.5
Eqs.
Solving
x.
wave
at
reflected
the
to
terms with subscript r refer
and reflected
simultaneously, we obtain explicit expressions for the incident

The terms with the

subscript

waves

^*[K(*) + Z,/(*)]
Vre*x = h[V(x)-Z

Consider the case when a transmission line of length

its characteristic impedance, that is,

V(L)

ZtfQ

(146)

l(x)]

is

terminated in

(14.7)

The

Then we

see that the reflected

wave

scattering matrix

415

is zero.

f>L

(14.8)

yL

cannot be zero, we see that the coefficient VT is identically zero

As a result, the reflected wave at any point x is zero. Also,
the impedance at any point x down the line is equal to Z as seen from
Eq. 14.5 with Vr = 0. With these brief thoughts of transmission lines in
mind, let us turn our attention to the main topic of this chapter, namely,
Since e

for this case.

the scattering parameters.

I4J

THE SCATTERING PARAMETERS FOR A

ONE-PORT NETWORK

For the one-port network shown in Fig. 14.2a, consider the following
The incident parameter a is defined as

definitions.

(14.9)

and the

reflected parameter b, is defined as

(14.10)

where R^

is

an

arbitrary, positive,

dimensionless constant called the

For the transmission line described in

Section 14.1, if the characteristic impedance Z = /{<,, then we can describe

reference impedance factor.

the incident parameter in terms of the incident voltage as

a
Similarly,

Ve~T"
- ^=-

(14.11)

b can be expressed in terms of the

b

reflected

wave as

= ^=r

(14.12)

VZ
/

One-port
network

(a)

a *
,

One-port
network

FIG. 142. Scattering parameters of a one-port network.

Network

416

Thus we
reflected

To

analysis

and synthesis

a and b do indeed describe an incident-

see that the parameters

wave

relationship in a one-port network, as depicted in Fig. 14.26.

give further

meaning to a and

b,

consider the power dissipated by the

one-port network

P = iRe

VI*

(14.13)

where /* denotes the complex conjugate of /. From Eqs. 14.9 and 14.10
we solve for V and / in terms of the incident and reflected parameters to

give

K=(a + b)V

Then the power

dissipated

(1414)

by the one-port network

P = Haa* -

is

bb*)
(14.15)

-KM* -1*1*)
where, again, the asterisk denotes complex conjugate. The term Jaa*
can be interpreted as the power incident, while $bb* can be regarded as the

power reflected. The difference yields the power dissipated by the one-port
network.

The incident parameter a and the reflected parameter b are related by the
equation

where

S is

coefficient.

= Sa

(14.16)

more commonly, the reflection

and b we can make the following

called the scattering element or,

From

the definitions of a

substitution:

Solving for S,

where

we obtain

Z is the impedance of the

= Z ~ **
Z+R
one-port network

Z=

excited

(14.19)

A further useful result is that when the impedance R<, is

impedance Z, the reflected parameter b 0.
For the one-port network

(14.18)

by a voltage source

set

equal to the

V9 with a source

The

we
when we choose the
impedance Rq to be equal to

resistor R,, as depicted in Fig. 14.3,

will

show

reference

AAA

"VW

that

One-port
network

!>

-&
The proof

By

follows.

incident parameter a

is

(,4 20>
-

definition, the

FIG. 14.3

given as

^)
-K+^)
From

Fig. 14.3,

we have

V,

and

-IRB

we

(14.21)

=V

(14.22)

(14.23)

Ra

Substituting these equations for

14.21,

417

scattering matrix

+Z

V and / into the expression for a in Eq.

obtain

_
_

VBR,

17.,

2V^

RqV,_~\

\
,

r r,+z

^+ zJ
(14.24)

R' + zJ

2VV

in a oneConsider once more the expression for the power dissipated

network:
port
V
(14.25)
P = l(aa* - bb*)

Factoring the term aa*

\a\*

from within the parentheses,

2 \

becomes

|a|V
(14.26)

=
When we
ance,

i.e.,

l^- (i-isi)

choose the reference impedance to be equal to the source

and
R, then S =
when R*

resist-

n _ nil!
~Z~ ~ on
2

ePA. ~

iR

(14.27)

Network

418

analysis

and synthesis

**

PA

where

~vVA

Av

>
^

*"

^*^

represents the available gain or

power of a voltage source Va with

a source resistance R. For the case of a

available

one-port network, the available gain is

defined as the power dissipated in the oneport network when the impedance of the

network
FIG. 14.4

Z is

equal to the resistance of the

As a result of this definition, we

source/?,.

see that for the one-port

Fig. 14.4, the

power

dissipated in

Pa

with

Z = Rt

network shown in

is

= ^-

The

available gain thus represents the

terminals of the voltage source.

(14.28)

maximum

available

power

at the

From this discussion, it is apparent that the value of the reference impedance Rq should be chosen equal to the source impedance R. A
standard procedure is to assume a 1-Q source impedance and denormalize
when

necessary, that

is, let

where
Next,

us

let

(14.29)

(14.30)

some of the important properties of the

a one-port network.

briefly consider

scattering parameter
1.

- 2^-i
z + 1

S for

The magnitude of S along theyco

for a passive network, that

axis

is

always

less

or equal to unity

is,

|S(/>)| <. 1

(14.31)

This property follows from the fact that the power dissipated in a passive
network is always greater or equal to zero. Since the power can be
expressed as

P-^fl-ISft^O
we

see that

2.

For a

\S(ja>)\*

(14.33)

network |S(jo>)| = 1. This property follows from the

power dissipated in a purely reactive network is zero.

reactive

fact that the

(14.32)

The
3. For an open circuit 5=1, and for a short
shown to be true from the equation

S
For an open
therefore

Before

5=

circuit z

oo,

circuit

S=

1.

= ^|
z + 1

so that

419

scattering matrix

This

is

(14.34)

5=1.

For a short

circuit z

= 0;

1.

we proceed

to the next property, let us consider the following

definition.

DEFINITION.

bounded real function P(s)

defined by the con-

is

ditions

|^)|

()

(6) F(s)

In

^ AT

is real

Rea^O

for

when

s is real.

K denotes any positive real constant.

(a),

4. If z

Z/Rq

a positive

is

real function, then

is

a bounded

real

function.

The proof follows from

S
From

the equation,

= (z-

l)/(z

the positive real condition (a)

Re z(s)

1).

^ 0, when Re s ^ 0,

we

see

that

s
so that

when Re j

|S(s)|

;> 0. (b)

Im z(s)-[l- Re z(s)]
jlmz(s)+ [I + Rez(s)]
j

+
= ( Im'^
llm , z(s) +

[l

Re2(S)]/

'

3 f

Where s is real, z(s) is real. Then

S = (zmust be real. Thus the

bounded real function.
14.3

U-Re^ *

(14.35)

l)/(2

1)

scattering parameter for a passive

network

is

THE SCATTERING MATRIX FOR A

TWO-PORT NETWORK

In this section we will extend the concepts developed for one-port

networks discussed in Section 14.2 to two-port networks. In the two-port
network shown in Fig. 14.5, we are concerned with two sets of incident and
reflected parameters {at , *x } at the 1-1' port, and {a,, b } at the 2-2' port.
t
These parameters are defined in similar manner as for the one-port

420

Network

and synthesis

anal/sis

-o

lo>-

-*o2

Ol

o-

02

v2

Two-port
network

Vi

*i

62

-o

FIG.

network, that

D2'

l'o-

14.5. Scattering

o-

parameters for a two-port network,

is,

(14.37)

"

=
l(vfe

where

R01

and

i?

+vr"'')
and output

are the reference impedances at the input

ports respectively.

The

scattering parameters

Stj for

the two-port network are given by the

equations

bt

(14.38)
iSjitfi

Si2a2

In matrix form the set of equations of Eqs. 14.38 becomes

6i

Si1

^im

(14.39)

(14.40)

where the matrix

is

called the scattering matrix of the two-port network.

From

Eqs. 14.38,

we see that the scattering parameters of the two-port network can be

expressed in terms of the incident and reflected parameters as

Sn
S

Si.

=h
a* 010

SM

=
a* ai0

(14.41)

The

scattering matrix

In Eqs. 14.41, the parameter

5M

is

Su

is

called the input reflection coefficient;

the forward transmission coefficient;

and SM

coefficient;

is

421

Slt

is

the reverse transmission

the output reflection coefficient.

Observe that

all

four scattering parameters are expressed as ratios of reflected to incident

parameters.

Now let us examine the physical meaning of these scattering parameters.

First, consider the implications

of setting the incident parameters at and a t

to zero in the defining relations in Eqs. 14.41. Let us see what the conimplies in the definition for the forward reflection coefficient
dition a4

tlt^O

Figure 14.6 shows the terminating section of the two-port network of Fig.
14.5 with the parameters a, and b a of the 2-2' port shown. If we treat
the load resistor

Rt as a one-port network with scattering parameter

S8

where Rgx

is

(14.42)

Ro

the reference impedance of port 2, then oi

and b% are

related

by*

aa

= SA

When the reference impedance R^

St becomes

is set

(14.43)

equal to the load impedance Rt,

then

S,=

Rot
Rat

Ri*8 =
+ Rto*

(14.44)

under this condition. Similarly, we can show that, when

so that at
ax
0, the reference impedance of port 1 is equal to the terminating

From

reflected

the viewpoint of the load resistor /? the incident parameter

parameter is a x .

is

b, and the

Network

422

impedance

and synthesis

anal/sis

R01 =

We see

as a result of this discussion that the

merely imply that the reference impedances
are chosen to be equal to the terminating resistors Rt and R t ,

(i.e.,

conditions a x

R01 and /?M

/?i).

and aa

respectively.

Next,

let

us consider the relationship between the driving-point imped-

ances at ports

and 2 and the

reflection coefficients

denote the driving-point impedances at ports

Let us

(14.45)

-M
a

the equation

5U and Su

and 2 as

h
From

(14.46)

l lai-0

We can write

=
*K W*oi)

which reduces

Similarly,

easily to

Sn

we have

These expressions

tell

us

if

Rqi
Z + Rqx
z Rq
Z + R oi

(14.47)

Rnli)

2q

(14.48)

Rt-Rot

h,_h

we choose

(14.49)

the reference impedance at a given

port to equal the driving-point impedance at that port, the reflection

coefficient

of that port will be zero, provided the other port

is

terminated

in its reference impedance.

Next, let us derive some physically meaningful expressions for the

forward and reverse transmission coefficients Sn and Slt Consider the
.

definition for <S81

321

-*l

(14.50)
a

implies that the reference imseen, the condition a, =

pedance Roi is set equal to the load impedance Rt , as seen in Fig. 14.7.
If we connect a voltage source Vgl with source impedance Rn = Rlt

As we have just

JtogaJfa

FIG. 14.7

The
then

we can

express

ax as
< 14- 5I >

""vfe
Since a,

423

scattering matrix

= 0, we have the equation

y
^
=

from which we obtain

bt

Consequently,

>/*'

(14.53)

= "(-/= - V*z
(14.54)

v*
Finally,

we can

express the forward transmission coefficient as

vjj**_

(14.55)

Krl^ ^* Ri-fi.i,Ri-Boi
In similar fashion, we find that when port 1 is terminated in Rqi = R t and
when a voltage source V,t with source impedance Rt is connected to port 2,

then

"-

?MF

(1456)

We see that both Slt and SM have the dimensions of a voltage-ratio transfer
function. Indeed, if
It is

R^ =

R^, then

Slt and Stl are simple voltage ratios.

Sn = Slt

seen that for a passive reciprocal network,

Now let us consider as an example the scattering matrix of the n: 1

ratio

Recall that for an ideal transformer

ideal transformer in Fig. 14.8a.

'Vi-jtV*

A---/,
n

(14.57)

Assuming first that Jin' Rn

1, let us find 5U by terminating the
2-2' port in a l-Cl resistor, as shown in Fig. 14.86. Then

-^-

(14.58)

Network

424

anal/sis

and synthesis

_Jf_
1:1|

2o
+

V2
Ideal transformer

(a)

FIG.

14.8.

Determination of scattering parameters for an ideal transformer,

From Eq.

14.48,

we have

Next we terminate the

as

we

^ = n*

Z1 =

so that

did for

Su

Sn

(14.59)

= n*-l
n* + l

1-1' port in

a 1-G

(14.60)

resistor (Fi. 14.8c).

We obtain,

SM

= (1/w*)
(!/*)

-1_1+1 1+

n*

(14.61)

The

425

scattering matrix

We obtain Sn by connecting a voltage source Vgl with a source impedance

= Q at the 1-1' port and terminating the 2-2' port with a resistance
- * the equivalent circuit
jRo, = 1 ii, as seen in Fig. 14.8rf. Since V^h

jRo!

of the ideal transformer as seen from the voltage source as a 1-Ii impedance in series with an *-ohm resistance (Fig. 14.8c). Then Vx can be
expressed in terms of Vtl , as

K,=
Since

Va =

(14.62)

n8

Vjn, we have

Kin
Since

R^ = R^ =

Q,

(14.63)

n*

SM is
2F,

2it

n*

r i

(14.64)
l

We can show in similar fashion that

2n

(14.65)
1

Therefore the scattering matrix for the ideal transformer

v-1
n*

As a second example,

let

transmission line of length

shown in

Fig. 14.9. If we

given as

In
n*

(14.66)

l-n

In
M*

is

fl*

+l

us find the scattering matrix for a lossless

its characteristic impedance, as

terminated in

assume that

R^ =

!*

= Z* then the reflection

coefficients are

= ?oZlZo _ Q
z, + z

(14.67)

x^-
Zo

1'

FIG,

14.9. Lossless

transmission line.

Network

426

This result

anal/sis and synthesis

not implausible, because a transmission line terminated in

impedance has zero reflected energy. To determine Su ,
we terminate the line in Z at both ends and connect at the 1-1' port a
voltage source Vtl , as depicted in Fig. 14.9. Since the transmission line
has zero reflected energy, that is,
its

is

characteristic

then

From

the equation

Z> x

v,

= *>- ri

a2

=
(14.68)

~ 2 \*J vgl
Sn - 2e-*

(14.69)

Stl

we obtain

we

In similar fashion,

(14.70)

find that

Sit

2e- rL

(14.71)

Therefore the scattering matrix for the lossless transmission line

r
~~

14.4

is

2-" z, i

(14.72)

VW-

PROPERTIES OF THE SCATTERING MATRIX

Having defined the

Section 14.3,
matrix.

let

From

scattering

matrix of a two-port network

in

us consider some important properties of the scattering

the general restriction for a passive network that the net

power delivered to

P=

all

we

ports must be positive,

Hii*

a*fh*

obtain the condition

W)

(14.73)

Equation 14.73 follows from the fact that the power delivered to the 1-1'
port

is

Pi

KW -

and the power delivered to the 2-2' port

P,

The

total

power

i(^*

delivered to the network

(14.74)

bib,*)
is

W)
is

(14.75)

then

P = P1 + Pt
which is exactly the expression in Eq.
delivered to the network is

(14.76)

14.73. In matrix notation, the

P - Hla'fW ~

lb*flb]}

power

(14.77)

The
where

T denotes the transpose operation,

lb]

[S][a],

then
[b*]

Equation 14.77 can

now

IP

=
-

[a*]

427

and

-a
-a

[a]

Since [b]

scattering matrix

(14.78)

T [S*] T

(14.79)

be rewritten as
{[a*]

T [a]

la*] llu]

- [a*] T [S*] T [S][a]}

- [S*] T [Smal >

This then implies that the determinant of the matrix

must be greater or equal to zero, that is,

Dct

[[]

(14.80)

[[]

[5*] T [S]]

[S*] [S]]

(14.81)

Consider the special but, nevertheless, important case of a lossless

P 0, so that

network. In this case

[S*] T [S]

[u]

(14.82)

A matrix satisfying the condition in Eq. (14.82) is unitary.

For a

lossless

two-port network

From

this equation,

we have

the following conditions for the scattering

matrix

Sj*Su

+ Stl *Stl =
+ Sn'Su =

Sn *Si,

+ Sn *S =

Su*Su

Su*S11

Note that Eqs.

network

14.85

is reciprocal,

and
then

S*S

(14.84)
(14.85)

(14.86)
(14.87)

14.86 are conjugates of each other.

If the

Stl = Slt and

+
IVHI +
|Sii(/>)|

=
|5(/))| -

|S(/>)|

(14.88)
1

428

Network

analysis and synthesis

Two-port
network

<*2

FIG. 14.10

from which

it

\SJ(ja>)\ <. 1
(i.e.,

=
=

follows that for a lossless reciprocal network |Su (ya>)|

1. Also it is clear that when \Sjj<o)\

and \Su(ja>)\

there is a zero of transmission), then \Su (ja>)\

all the power that has been delivered to the

when

condition states that

1.

This

network

back to port 1-1'.

At this point, it might be profitable to discuss why we use scattering
matrices. What are the advantages of the scattering description over
conventional descriptions? Let us discuss three major reasons for the

from port

1-1'

is

reflected

scattering formalism.

networks do not possess an impedance or admittance matrix.

ideal transformer has no Z or Y matrix because its
elements are not finite. However, as we have seen, the ideal transformer
8
can be described by a scattering matrix. Carlin states that all passive
1.

Many

For example, an

networks possess scattering matrices.

2. At high frequencies, incident and reflected parameters play dominant
roles in problems of transmission, while voltage-current descriptions are
relegated to the background. Then the scattering matrix is necessarily the
more powerful description of the system. Note that the voltage standingwave ratio (VSWR) is given in terms of a reflection coefficient S as

VSWR =

iM

(14.89)

In networks where power flow is a prime consideration (e.g., filters),

the scattering matrix is very useful. For example, in the network given in
Fig. 14.10, ifPA represents the available power from the generator and Pt
is the power dissipated in the load R t , then we can.show that the magnitude3.

squared forward transmission coefficient

is

iso)i*-*
pA

We will discuss this point in more detail in

< 14 - 90>

Section 14.5.

* H. J. Carlin, "The Scattering Matrix in Network Theory," Trans. IRE, CT-3,

No. 2, June 1956, 88-96; see his extensive bibliography.

The

429

scattering matrix

INSERTION LOSS

14.5

we described the forward and reverse transmission

terms of voltage ratios. Perhaps a more appropriate

In Section 14.4,
coefficients in

description of a transmission coefficient

is in terms of a power ratio rather

than a voltage ratio. In this section we will show that |SM(/o)|* and ]SU
(J(o)\* can be expressed in terms of power ratios. We will then introduce
the very important concept of insertion loss and finally relate |ti(/<)l* to

the insertion power ratio.

Consider the equation for

Stl

in the two-port

network shown in Fig.

14.5,

From
we

the equation

obtain

l-S^OOl*

= Sn(/w)Sa *(/a>)

|M-)|-

(14.92)

l^>
I^iO)IV8*x
(14.93)

Pal
where Pt is the power dissipated by the load Rt, and
gain of the generator Vgl Similarly, we have

PAl is

the available

|Si.O)l*

We

see that both \Su (j(o)\*

relate the

power dissipated

and

- -

ISutyco)!* are

(14.94)

power

transfer ratios

at a given port to the available

power

which
in the

other port.

Now

us examine the idea of insertion

loss. Consider the network

Between the terminals 1-1' and 2-2' we will insert a
two-port network, as shown in Fig. 14.116. Let us denote by VM the
voltage across the load resistor R, before inserting the two-port network,
and by Vt the voltage across Rt after inserting the two-port network. A
measure of the effect of inserting the two-port network is given by the
insertion voltage ratio IVR, which is denned as

let

shown in

Fig. 14.1 la.

IVR

^V

(14.95)

is

Another method of gaging the effect of inserting the two-port network

power dissipated at the load before and after inserting the

to measure the

430

Network

analysis

and synthesis

-J

vW^-

r
(a)

Hi

-i_

r
Inserted

)yi

*2* >V2

two-port
network

r
lb)

FIG. 14.11

two-port network. If PM is the power dissipated at the load before the

two-port network is inserted, and if Pt is the power dissipated after insertion, then the insertion power ratio of the two-port network is denned as
e

If

we

_ Ew

take the logarithm of both sides,

a
where a

is

(14.96)

we

=10 log

obtain

(14.97)

the insertion loss of the two-port network. In terms of the

we can calculate Px from the relation

circuit given in Fig. 14.11,

Then

Pm is

|K|

dissipated

(14.98)

2R t
*.
2(R t

The power

R>

by the load

ir.il'

(14.99)

RJ*

after inserting the two-port

network

is

given by

P.= Hi!!
2R t

(14.100)

The

The

insertion

power

ratio

2a

when

431

can then be expressed as

_80

_ wr

P.-VXOk +
In the special case

scattering matrix

<14101)

the source and load impedances are equal, that

is,

Rx = R2 = Rn = R^

(14.102)

the reciprocal of the squared magnitude of the forward transmission

coefficient in Eq. 14.93 is equal to the insertion power ratio
Pgo

When Rt

j&

Rt

|S*iO)|
,

then
PjQ

4R1 R 2
(*i

In any event,
i\Sa (jai)\ t

power

we

(14.103)

1
2

R*) |s*iO)l

(14.104)

see that the magnitude-squared transmission coefficients

and {S^jai)]* can be regarded physically as equivalent

insertion

In Section 14.6, we will use this relationship in the synthesis

of double-terminated filter networks.

14.6

ratios.

DARLINGTON'S INSERTION LOSS FILTER SYNTHESIS

In this section,

we will consider a filter synthesis procedure first proposed

classic paper in 1939. 4 We will use scattering matrix

notation to describe the essence of Darlington's original work. Our
coverage will be restricted to the class of low-pass filters which are termi-

by Darlington in a

nated in equal source and load impedances, Rn

Roi R , as shown in
For normalizing purposes we will let /^ be equal to 1 12.

Fig. 14.12.

FIG. 14.12

S. Darlington, "Synthesis

Loss Characteristics,"

/.

of Reactance 4-Poles which Produce Prescribed Insertion

Math. Phys., 18, 1939, 257-353.

Network

432

analysis and synthesis

Recall that when the source and load

impedances are equal, then the insertion
power ratio is equal to the reciprocal of
|SM (yG>)| 2 , that

is,

20

(14.105)

|S21 (>>)| a

P*

Expressed as a loss function, the insertion

power

ratio is

A =10 log %*
P

-101og|S 21(| a

db

(14.106)

In circuit design, the specification of an

A (Fig. 14.13a) is equivalent

insertion loss

to the specification

squared transmission coefficient shown in Fig.

of the

14.13ft.

amplitude-

One of the most

ingenious techniques given in Darlington's synthesis procedure is the reduction of insertion loss synthesis to an equivalent L-C driving-point
synthesis problem. This technique can be developed in terms of scattering
parameters. Our initial specification is in terms of |Su (/<u)|. For an L-C

two-port network

\Sn(jcoW
Next,

Sn(s)

is

the equation

Su

=1-

\S ia (Jo>)\

=
+

^i

we

|S*iO)l

(14.107)

obtained from the magnitude-squared function

Su (s)Sii(- s)
Then from

=1-

2
\

im ^,

(14.108)

(14.109)

obtain the driving-point impedance

Z^s)

=R

+ Su(s)
- Su (s)

(14.110)

We then synthesize the network from Z^s).

discussion here to low-pass filters given by the
our
We
lossless ladder structure terminated at both ports by 1-Q resistors in
Butterworth or
Fig. 14.14. These low-pass filters can take the form of a
shown

in Fig. 14.12.

will restrict

Chebyshev

specification for |Ssl (/a))|

|S21

2
,

that

=
0)f _
1 +

is,

(14.111)
,2n

CO

The
10

scattering matrix

433

Li
/-

~*np

T-nnnp

c2 z

T-

C-1-

FIG.

14.14.

Canonical form for double-terminated low-pass

where

10.

Zi(s)

niters.

C B8(<u)

v2

(14.112)

Cn(co) represents an nth-order Chebyshev polynomial.

Example

14.1.

Let us synthesize a low-pass

filter

for the specification

._.
"*W-TT^

(14.113)

which represents a third-order Butterworth amplitude characteristic. The load

and source impedances are R^ = R^ = 1 Q. First we find |5u(y'o>)| s as

IWI* = !--*
1

Lettingy'eo

= s in

|Su

(/a>)| a

we

(14.114)

+0)6

obtain

<Sii(>S_(-*)

= 1

-s*

(14.115)

which factors into

Sn(s) Sil(- S) sothat5u(5)is

Next,

8
A-*
)
"

g +2> +

^ + ^ 1^ +2j2 _ ^

_______

(14.116)

a*

sll(f) =

(14 117)
.

Zx(j) is obtained from the equation

Zfy)

+ Su(s)
- SuM

2s*

+2s* +25
2s?

2s

(14.118)
1

Network

434

analysis and synthesis

FIG. 14.15

We next perform a Cauer ladder expansion for Z^s).

2s*

+ 2* + l)^ + 2s + 2s +

l(s

2s*+2s*+s
s

1)2**
2s*

+2s +

l{2s

Zs
1)5

The low-pass

filter is

An equivalent
we

realization for the double-terminated filter is obtained if

Su(5)

= YM - G

G = 1 mho
1
1

14.2,

and

14.3

(14.119)

Yt(s) + G

Y1(s) =
The canonical

1(5

thus synthesized in the structure shown in Fig. 14.15.

use the equation

Then, assuming

+ Sn (s)
- Su(s)

realization for Yx(s) is

list

shown

element values (up to n

(14.120)

in Fig. 14.16.

Tables 14.1,

7) for double-terminated

Butterworth, Chebyshev (1-db ripple) and Bessel filters, respectively.

These apply to the canonical realization for Yt(s) given in Fig. 14.16. If
a ZjCs) realization in Fig. 14.14 is desired, we simply replace all shunt
capacitors by series inductors

FIG. 14.16.

and

vice versa.

Canonical form for

filters

in Tables 14.1, 14.2,

and

14.3.

The

TABLE

scattering matrix

435

14.1

Normalized Element Values for a Double-Terminated

Butterworth Filter (Equal Terminations)
n

Ci

2.000

Lt

C3

Lf

Lg

Cj

C?

1.414

1.414

1.000

2.000

0.765

1.848

1.848

0.618

1.618

2.000

1.618

0.618

0.S18

1.414

1.932

1.932

1.414

0.518

0.44S

1.248

1.802

2.000

1.802

1.248

1.000

0.765

TABLE

0.445

14.2

Normalized Element Values for a Double-Terminated

Cheb/shev Filter with -decibel Ripple (Equal Terminations)
I

71

Cx

L*

C8
2.024

c5

L*

1.018

2.024

0.994

2.135

1.091

3.001

1.091

2.135

2.167

1.112

3.094

1.174

3.094

TABLE

J-6

c7

1.112

2.167

14.3

Normalized Element Values for a Double-Terminated

Bessel Filter (Equal Terminations)

C3

La

Cj

C3

Lt

Le

Cj

2.000

1.577

0.423

1.255

0.553

0.192

1.060

0.512

0.318

0.110

0.930

0.458

0.331

0.209

6
7

0.838

0.412

0.316

0.236

0.148

0.051

0.768

0.374

0.294

0.238

0.178

0.110

Note that the even orders

are not given. This

is

0.072

for the double-terminated

0.038

Chebyshev niters
filters do not

because the even-ordered Chebyshev

meet realizability conditions for minimum insertion loss at s = 0. 5 We

have only given tables for equal source and load terminations. For other
possible realizations, the reader should consult L. Weinberg's excellent

book. 8

L. Weinberg, Network Analysis

New York,
'Ibid.,

1962, p. 589.

Chapter

13.

and

Synthesis,

McGraw-Hill Book Company,

Network

436

anal/sis

and synthesis

Problems
14.1 Determine the reflection coefficient
in the figure.

S for

the one-port networks

shown

?o

-VWrvW
4=tt'

w
For the one-port_network in Fig. 14.3, let iJ, . R
If the incident
B
is a = VJl^R^ find the reflected parameter
A.
14.3 For the network in Prob. 14.1, determine
\S(jio)\.
Show that the
scattering elements S for the networks in Prob. 14.1 are
bounded real functions.
14.4 For each of the networks shown, find the
scattering matrix for ik, =
14.2

parameter

-o2

(a)

(b)

+
Vi

/2-*-

^-J-?-

Gyrator

/i

A
V

Negative

impedance
converter

*!,

^ W*
*1

Vi

w
PROB.

14.4

1.
z

T
w

v2

The

scattering matrix

437

14J5 Find the insertion voltage ratio and insertion power ratios for each of
the networks shown. These networks are to be inserted between a source
impedance R = 2 ft and a load impedance RL = 1 ft. From the insertion

power

ratio, find |Si(j>)|*.

2h_
i

2_h

o-'TRRTH-^WnP-02
Jsjf

io

_lh_
.

nptpP 1

=f=if

=r=i'

l'o-

(6)

2h

40
:*f

PROB.
14.6

Synthesize low-pass niters for the specifications

(a)

|SaQ'>)l*

l-W)!

=
1

+0)*

+afl

Synthesize an equal-ripple low-pass filter such that 20 log |5n(/a>)| has

most i-db ripple in the pass band and an asymptotic falloff of 12 db/octave in

14.7
at

14.5

the stop band.

chapter 15

Computer techniques

in

circuit analysis

15.1

THE USES OF DIGITAL COMPUTERS

IN

CIRCUIT ANALYSIS
The advent of the high-speed computer has made

routine many of the

computational aspects of circuit theory.
Digital computers have become widely used in circuit analysis, time and
frequency-domain analysis, circuit (filter) design, and optimization or

formerly tedious and

We

iterative design.

difficult

will discuss these aspects in general in this section.

In succeeding sections,

we

will

discuss

some

specific

circuit-analysis

computer programs.
Circuit analysis
The primary objective of a linear circuit-analysis program is to obtain
responses to prescribed excitation signals. These programs are based on
many different methods: nodal analysis, mesh analysis, topological
formulas, and state variables. Most of them can handle active elements
such as transistors and diodes by means of equivalent circuit models.

The

state-variable

programs based upon Bashkow's

tion 1 perform their calculations directly in the time

matrix formula-

domain

via numerical

and matrix inversion. The outputs of these programs provide

impulse and step response in tabular form. If the excitation signal were
given in data form, the state-variable programs would calculate the
response directly in the time domain.
integration

T. R. Bashkow,

Theory, CT-4, No.

"The A MatrixNew Network Description," IRE Trans, on Circuit

3,

September 1957, 117-119.

438

Computer techniques

The majority of

circuit-analysis

in circuit anal/sis

439

programs, however, perform their

calculations in the frequency domain. The program user is only required

to specify the topology of the network, the element values, and what
transfer functions he wishes to obtain. The computer does the rest. It
calculates the specified transfer functions in polynomial form, calculates

the poles and zeros of these functions, and can also provide transient
response and steady-state response, if desired. With versatile inputoutput equipment, the output can also provide a schematic of the original

network, as well as plots of time- and frequency-response characteristics.

Time- and frequency-domain analysis

The time- and frequency-domain analysis programs can be used in
conjunction with the circuit-analysis programs or independently. The
time-domain programs depend upon solving the convolution integral

FIG. 15.1. Frequency response of fifth-order Butterworth filter.

FIG. 15.2. Phase response of fifth-order Butterworth filter.

440

FIG.

Network

15.3.

analysis

and synthesis

Impulse response of fifth-order Butterworth

filter

evaluation by Laplace

transform.

This approach obviates the necessity of finding roots of

It has the advantage that the excitation signal
need not be specified analytically, but merely in numerical form.
The frequency-domain programs usually consist of finding transient
and steady-state responses, given the transfer function in factored or
numerically.

high-order polynomials.

The program user must specify the numerator and

denominator polynomials of the transfer functions, the types of transient
response he wishes (i.e., impulse or step response), and the types of
steady-state responses he wishes (amplitude, amplitude in decibels, phase,
delay, etc.). In addition, he must specify the frequency and time data
points at which the calculations are to be performed. This may be done
in two ways. If he requires evenly spaced data, he need only specify the
minimum point, the increment, and the number of points. If he wishes
unfactored form.

to obtain data at certain points, he

must supply the

list

of data points at

the input.

Examples of outputs of a steady-state and transient analysis computer

program are shown in Figs. 15.1, 15.2, and 15.3. In Fig. 15.1, the magnitude of a fifth-order Butterworth filter is plotted via a microfilm plotting
subroutine. Figure 15.2 shows the phase of the filter, while Fig. 15.3
shows its impulse response.

Computer techniques
In Section 15.2

we

examine further

will

in circuit analysis

details

441

of a typical steady-state

analysis program.

Circuit

The

(filter)

filter

design

design programs are probably the most convincing argument

for the use of computers in circuit design. Designing insertion loss filters"
to meet certain amplitude requirements requires considerable numerical
calculation even in the simplest cases.
insertion loss

filter

design

is clearly

The use of

digital

logical alternative.

computers in

The amount of

programming time for a general filter synthesis program is considerable.

However, the ends certainly justify the means when large numbers of
filters must be designed to meet different specifications.
An outstanding example of a digital computer program for filter design
The
is the one written by Dr. George Szentirmai and his associates.*
complete in that it handles the approximation as well as the
synthesis problem. It is capable of dealing with low-, high-, and band-pass
poles
filters with prescribed zeros of transmission (also called attenuation
maximally
or
equal-ripple
either
for
or loss peaks). There are provisions
flat-type pass-band behavior, for arbitrary ratios of load to source im-

program

is

pedances, and for predistortion and incidental dissipation.

In addition, Dr. Szentirmai has a modified program that synthesizes
low- and band-pass filters with maximally flat or equal-ripple-type delay
in their pass band, and monotonic or equal-ripple-type loss in the stop

bands.
In the specifications, the designer could specify both the zeros of
4
If
transmission (loss peaks) and the network configuration desired.
his specifications include neither, the computer is free to pick both configuration and zeros of transmission. The computer's choice is one in

which the inductance values are kept at a minimum. The program was
written so that the same network could be synthesized from both ends.
Finally, the computer prints out the network configuration, its dual,
the normalized element values, and the denormalized ones. It also
provides information such as amplitude and phase response, as well as
plots of these responses obtained from a microfilm printer.
Figures 15.4, 15.5, 15.6, 15.7, 15.8, and 15.9 show the results of a
band-pass

filter

synthesis using Dr. Szentirmai's program.

R. Saal and E. Ulbrkh,

Circuit Theory, CT-5,

"On

December

the Design of Filters

by

Figure 15.4

Synthesis," Trans.

IRE on

1958, 287-327.

Digital Computer Program Package for

of the First Allerton Conference on Circuit and System
Theory, November 1963, University of Illinois.

G. Szentirmai, "Theoretical Basis of a

Filter Synthesis," Proceedings

4

Saal and Ulbrich, op.

cit.

Network

442

analysis

and synthesis

BAND PASS FILTER SYNTHESIS

CASE NUMBER

3.1

OEGREE OF FILTER
MULTIPLICITY OF PEAK AT ZERO
MULTIPLICITY 0F PEAK AT INFINITY
NUMBER OF FINITE PEAKS BELBM THE BANO
NUMBER #F FINITE PEAKS ABBVE THE BANO
EQUAL RIPPLE PASS BANO REQUESTED
PASS BANO RIPPLE MAGNITUDE
L0HER PASS BANO E06E FREQUENCY
MID-BAND FREQUENCY
UPPER PASS BANO EDGE FREQUENCY

13
3
2
1

0.09000
DB.
1.0000000E+04 CPS.
1.3416408E*04 CPS.
i.80oooooe*04 cps.

"ARBITRARY" STOP BANO REQUESTED

NUMBER 0F SPECIAL POLES

l.OOOOOOOE+00

MA

TERMINATIONS
INPUT TERMINATION
OUTPUT TERMINATION

6.0000000E02 OHMS
0.
OHMS

LOWER FINITE STOP BANO PEAKS

FREQUENCY CPS
NORMALIZED
6.2000000E+03
4.621 2071E-01

VALUE OF M
0.6229266

UPPER FINITE STOP BANO PEAKS

FREQUENCY CPS
NORMALIZED
2.0700000E+O4
1.5428869E+00
2.31000006*04
l.7217723E*00
3.5200000E+04
2.6236531E+00

VALUE OF H
2.3788111
1.9296560
1.4968796

COMPUTER HILL SPECIFY CONFIGURATION

LAST INDUCTOR IS A SERIES BRANCH
FIG. 15.4

gives the specification of the problem.

The pass-band magnitude

is

equal ripple with ripple magnitude of 0.05 db, and the degree of the
is

As we

to be 13.

to be
filter

indicated in Chapter 14, odd-degree, equal-ripple

are nonrealizable. In this example the designer utilized an ingenious

filters

an extra pole to accomplish the synthesis. The program logic

then provided two extra zeros: one to cancel the extra pole and the other
to provide the odd degree. The extra pole is called a special pole in Fig.
1.0.
15.4, and is located at s
Further specifications call for the lower band-edge frequency to be 10*
device

cycles;

the upper, 1.8

Vl.8 X

10*

1.3416408

zeros of transmission

10* cycles;

at/=

10* cycles.
(three),

and the midband frequency to be

In the stop band, there are to be

/=

oo (two),

and four

finite

zeros

of transmission: one below the pass band and three above the pass band.
The positions of these finite zeros of transmission are chosen by the
designer as indicated by the notation "arbitrary" stop band requested.

Computer techniques
CASE NUMBER

FMUtD

1.1

OtMEE OF FILTER

13

MM

EQUAL RIPPLE PASS SAND

LONER -PASS-OAM EOSE FREQUENCY
UPPER PASS-BANO EOCE FREQUENCY
MID-BAND FREQUENCY

1
2

0.0500

ARBITRARY' STEP

DB

cps
1.3416408E*04 CPS
i.aooooooE04

100

300

NORMAL IZEO

UNNORHALIZEO

1.0000000E*00

6.0000000E*02

3.0T04337E-01

6.0706102E-09
S.9892B16E-03
3.85TA228E-09

2.9723019E00

9. 089744 OE-08

1.0083380E00
4.44298O6E00

..L.e..:

7.1783955E-03
9.1797212E-08

TERMINATION
803

PEAK FREQUENCY
3.5200000E+04
901

PEAK FREQUENCY
4.2000000E03
801

2.6982703E-08

1.364T491E00

L C

9.3749272E-01
7.S1321S3E-01
7.9423894E-01

1.44M444E*00
i""c

2.T112T46E-01
t.2441S6*E00

3.S2A810TE-03
1.9447656E-0B

PEAK FREQUENCY
2.0700000E04

1.4931737E-08

100

2.8S46S14E-08

802

1.429T93TE-03
2.499S449E-08

PEAK FREQUENCY
2. 310000 0E04

1.0127M7E01

2.0022669E-07

400

9.97702ME-02

.0165648E-04

200
100

l.A3787S9EQ0

3.2382 800E-08

4.2943194E-01

4.4B148I6E-03

300

TERMINATION

SHORT

SNORT

(AND)

- i.oooooooso* cps

CONFIGURATION SPECIFIED QV COMPUTER

03 901 *01 100 002 *00 200

L*"e

443

REALIZATION FROM A SDMI CIRCUIT ADMITTANCE

MULTIPLICITY OF PEAK AT
MULTIPLICITY OF MAX AT INFINITY -

T.8921337E-01
1.8499769E-0I

in circuit analysis

FIG. 15.5

The terminations
filter

are:

input

60 Q, output

= 0Q

terminates in a short circuit. In this example, the

which means the

filter

configuration

inductance configuration. 8

chosen by the computer, and is a minimum

In Fig. 15.4 there are, in addition, listings of the finite zeros of transmission (loss peaks), which the designer specified.
Figure 15.5 is a printout of the configuration of the filter as shown by
the dotted lines flanked by the associated element values, both normalized
Since there are four
(left column) and unnormalized (right column).
finite zeros of transmission, there must be associated four L-C tank circuits.
is

W.

Saraga,

"Minimum

30, July 1953, 163-175.

Inductance or Capacitance Filters," Wireless Engineer,

Network

444

FREQUENCY
IN CVCt.ES
16000
16250
16300
16730
17000
17250
17500
17750
10000
10250
16500
16750
19000
19250
19500
19750
20000
20250
20500
20750
21000
21250
21500
21750
22000
22250
22500
22750
25000
25250

anal/sis

and synthesis

VOLTAGE RA
TIO IN 08

IN OEGREES

5.716532
5.704379
5.707644
5.729113
5.750516
5.743140
5.708979
5.720983
5.702947
4.131556
-1.119811
-7.753694
-14.148566
-20.210264
-26.118735
-32.100833
-38.473164
-45.657895
-56.325030
-71.701988
-59.444189
-57.483949
-57.567501
-58.703992
-60.546362
-63.101161
-66.602572
-71.867460
-63.403713
-80.597817

145.549717
128.958686
111.534102
92.934806
72.763652
50.659800
26.219204
356.194819
322.320795
272.913465
225.689589
195.922988
177.085862
163.748654
153.507803
145.227029
138.296917
132.354567
127.166241
302.572268
298.458582
294.740796
291.334706
288.250341
285.388013
282.735723
280.267304
277.961066
275.798866
93.765362

PHASE

NAG

REAL I IN
IN 8HNS

2 IN
IN 8HNS

134.691
130.664
125.883
119.931
112.687
104.489
95.717

REAL V IN
IN

56.981
61.193
64.762
67.928
71.154
74.696
77.919
79.138
78.767
87.140
111.450
139.624
164.004
184.583
202.605
218.952
234.147
248.501
262.211
275.407
286.178
300.592
312.696
324.536
336.138
347.527
358.726
369.753
380.621
391.345

65. 165

67.626
40.032
16.386
3.462
1.725
.541
.167
.049
.013
.002
.000

.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000

NILNMS
6.2973
6.2766
6.2814
6.3129
6.3445
6.3337
6.2835
6.3011
6.2748
4.3533
1.2913
-2797
.0641
.0158
.0040
.0010
.0002
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000
.0000

INA6 V IN
IN NILNHOS

-2.6641
-2.9395
-3.2315
-3.5756
-4.0061
-4.5277
-5.1151
-5.8552
-7.3083
-9.4759
-6.7826
-7.1517
-6.0967
-5.4176
-4.9357
-4.5672
-4.2708
-4.0241
-3.8137
-3.6310
-3.4701
-3.3268
-3.1980
-3.0813
-2.9750
-2.8775
-2.7876
-2.7045
-2.6273
-2.5553

FIG. 15.6

20

(
1

-20

-40

-60

-80

f\
1

/
/
'

-100
-120
0.000

0.400

Add 1.00000

0.800

1.200

1.600

2.000

03 to abscissa
Frequency, cycles

2.400

FIG. 15.7

10*

2.800

3.200

3.600

4.000

Computer techniques

445

in circuit analysis

--

JE
;
J

-1
0.000

0.400

Add 1.00000S

0.800

1.200

2.000

1.600

03 to abscissa
Fl6queflcy

2.800

2.400

3.200

3.600

4.000

x ,

FIG. 15.8

0.800

0.600

0.400

0.200

0.000

(*"

-0.200

-0.400

--

-0.600

'*

\
1,

-0.800

u
-1.000
0.000

0.400

Add 1.00000E

0.800

1.200

+ 03 to abscissa

1.600

2.000

}nmm

FIG. 15.9

2.400

10

2.800

3.200

3.600

4.000

Network

446

The peak

anal/sis

and synthesis

the

column entitled "termination,"

on the same line two columns to

frequencies are listed under the

and the associated L-C tank

circuits are

left.

is a listing of a small portion of the frequency response,

which was calculated after the filter had been synthesized. Figure 15.7

Figure 15.6

a plot of voltage ratio in decibels versus frequency.

Note that 1000

must be added to every frequency value in the abscissa. This
merely reflects an idiosyncrasy of the plot routine.
Figures 15.8 and 15.9 show the real and imaginary parts of the input
admittance of the filter. Note particularly the shapes of these characteristics. If a number of these band-pass filters were connected in parallel
and if the pass bands of the filters were adjacent but nonoverlapping,
then the input conductance of the filter system would be essentially
constant over the entire pass band, while the input susceptance would
cancel out, as shown in Figs. 15.10a and b. This then means that the
input admittance of the filter system would be real and could then be
driven by any arbitrary source impedance without fear of reflections.
Filters obtained using computer programs of the type we have just
described are rapidly supplanting conventional hand- and handbookdesigned filters. The filter synthesis programs provide filters that are
orders of magnitude more sophisticated than niters designed by the
is

cycles

Frequency
(a)

FIG. 15.10. Effects of paralleling several band-pass

filters

with adjacent pass bands.

Computer techniques

447

in circuit anal/sis

Moreover, the designs are completed in minutes

a typical cost of twenty dollars rather than two
thousand (not including initial programming costs, of course).
conventional manner.
rather than days

and

at

Optimization
Network design cannot always be accomplished by way of

analytical

Quite often networks are designed through a trial-and-error

process. The network designer begins with a set of specifications. He
then selects a network configuration, and makes an initial guess about
the element values. Next he calculates or measures the desired responses
and compares them with the specifications. If the measured responses
differ by a wide margin from the specified responses, the designer changes
the values of the elements and compares again. He does this a number
of times until (hopefully) the measured responses agree with the specified

means.

responses to within a preset tolerance.

made to converge, sometimes quite

one uses a method of steepest descent.* To get a rough idea
of the steepest descent method, suppose there are n parameters in the
network. Let us regard each parameter x{ as a dimension in an n,xn)
dimensional euclidean space. We define a function f(x u xa xs
that assigns a functional value to each point of the euclidean space. We
then ask, at point p t what direction of motion decreases the value of
f(xlt xit *,...,*) most rapidly? The function f(xt) may be defined as
a least squared error, or as an absolute error between calculated response
and specified response. The direction of steepest descent is the direction
denned by the gradient
This process of cut and try can be

rapidly, if

+ T-* +
mdf=T-*i
OX
OXi

---

+ ^*n
OX

Therefore, incremental value of change for each parameter

6xt

(15.D

is

=-C

(15.2)

OXi

where

is

a constant. After

all

incremental value in Eq. 15.2, a

are obtained.

An

parameters have been changed by the

gradient and new incremental values

new

The process continues until

by C. L. Semmelman7

excellent paper

the

optimum

is

obtained.

describes a steepest descent

* Charles B. Tompkins, "Methods of Steep Descent," in Modern Mathematics for the

Engineer (Edwin F. Beckenbach, ed.), Chapter 18, McGraw-Hill Book Company, New

York, 1956.
7
C. L. Semmelman, "Experience with a Steepest Descent Computer Program for
Designing Delay Networks," IRE International Convention Hec, Part 2, 1962, 206-210.

448

Network

analysis

and synthesis

Fortran program used for designing delay networks. The specifications

Rt given at the frequency data points/,. The program

are the delay values

successively changes the parameters

xi so

that the squared error

<*,)=i[K,-n*<,/,)]

(i5.3)

i-X

minimized. In Eq. 15.3, T{x it ft ) represents the delay, at frequency ff

of the network with parameters xt
In order to use the program, the network designer must first select the
initial values for the parameters x (
He must also provide the specified
delays R, and the frequency data points/^. The program provides for 128
match points and 64 parameter values. It is capable of meeting requirements simultaneously in the time and frequency domains. The designer
is not restricted to equal-ripple approximations or infinite Q requirements.
He is free to impose requirements such as nonuniform dissipation and
range of available element values on the design. For related methods of
optimization, the reader should refer to a tutorial paper by M. R. Aaron. 8
is

Machine-aided design
The concept of real-time

man and computer holds

network design. Multiple-access computing
systems, such as the project
of Massachusetts Institute of Technology, make cut and try design procedures practicable. The initials
could describe either the term multiple-access computer or the term
machine-aided cognition. In an article describing the
computer
system, 9 Professor R. M. Fano states "The notion of machine-aided cognition implies an intimate collaboration between a human user and a
computer in a real-time dialogue on the solution of a problem, in which
the two parties contribute their best capabilities."
A simple multiple-access computer system is shown in Fig. 15.11.
There are n data links and n terminals connected to the central processor.
Located at each terminal is input-output equipment, such as teletypewriters, teletypes, and oscillographic displays. The sequence in which
the central processor accepts programs from the terminals is controlled by
a built-in queueing logic.
The main reason a multiple-access computer is effective is that the
central processor computes thousands of times faster than the user's
reaction time. When a user feeds in a program, it seems only a "moment"
much promise

interaction between

in the field of

MAC

MAC

MAC

M. R. Aaron, "The Use of Least Squares in System Design," IRE Trans, on

December 1956, 224-231.

R. M. Fano, "The MAC System: The Computer
Spectrum, 2, No. 1, January 1965, 56-64.

Circuit

Theory, CT-3, No. 4,

Utility

Approach,"

IEEE

Computer techniques

in circuit analysis

Remote

Data

Memory

449

terminal

bank

link

Memory

Central

bank

processor

Buffer

Data

memory

link

Remote
terminal

Queueing

Data

logic

link

Remote
terminal

Memory
bank

Memory
bank
Data

Remote
terminal

link

FIG.

15.1 1.

Block diagram showing typical multiple-access computing system.

before he gets the results through the teletypewriter or oscilloscope

display. He examines the results, changes some parameters, and feeds the

program into the central processor again. The queueing time and processing time on the multiple-access computer may amount to only a
minute or two, which, for the user, is probably not significantly long.
Dr. H. C. So has written a paper on a hybrid description of a linear
10
in which he has shown that the hybrid matrix is ideally
ii-port network
suited for such problems such as multielement variation studies and
iterative design. Suppose there are n variable elements in the network.
These can be "extracted" and the rest the bulk of the network can be

described by the n-port hybrid matrix.

10 H. C. So, "On the Hybrid Description of a Linear i-Port Resulting from the
Extraction of Arbitrarily Specified Elements," Trans. IRE on Circuit Theory, CT-12,
No. 3, September 1965, pp. 381-387.

'

Network

450

analysis

and synthesis

Command

Initialize

i
1

Frequency
range
Engineer

Control

Computer

*j

Parameter

adjust

'

Display

Output

FIG. 15.12. Iterative design of network using machine-aided cognition.

Dr. So has written a computer program to formulate the hybrid matrix

for the H-port network automatically. 11

The inputs to this program are

node connections specifying the network topology; (2) the impedance functions of the elements; and (3) the specifications of the
special elements to be extracted. This program was written with manmachine interaction in mind. The process is described in Fig. 15.12.
First the computer reads in the n-port program with initial specifications.
It performs the calculations and feeds information, such as transient or
steady-state responses, back to the engineer via a visual display console.
The engineer assesses the data and then changes the parameter values of
(1) the

the extracted elements, and, in

some

instances, the frequency range over

which the calculations are made. The process

is

repeated a

number of

times until the engineer has obtained the desired results. Such a program

could also be controlled by a steepest descent steering program

if

the

engineer wished to obtain his results with less "eyeballing."

Now let us examine some details of specific network analysis programs.

15.2

AMPLITUDE AND PHASE SUBROUTINE

Purpose

Our purpose

is

to

compute the amplitude and phase of a

rational

function

"(s) - ?Hr?
1

H. C. So, unpublished memorandum.

(154)

Computer techniques

in circuit anal/sis

451

over a set of frequencies (o^ In Eq. 15.4, C and C are real constants,
d
and E(s) and F(s) are polynomials in s = a +ja>. The program described
here computes the amplitudes M(co,) and phase #a> ) over a set of
t
frequencies cok where
,

JU

W =

\CnE(jlOh)\

(15.5)

\Ca FOcok)\

and
<f>((ok)

= arctan

arctan

ft

Fj(ja> k)

(15.6)

where in Eq. 15.6 the

indicate

real

respectively.
<Ko>k)
<f>

*s

and

subscripts

and imaginary
Note that in Eq.

limited to the

<,it radians.

range

parts,
15.6,

it

<>

In order to circumvent

this restriction on the phase, we use the

following method to compute amplitude

-JW
FIG. 15.13. Calculation of magnitude and phase at a> ( due to zero z.

and phase.

Method used
Let H(s) be factored into poles and zeros such that

H(s)

= C n (s - z )(s - zt)
Cd(s ~ Pi)(s - P)
t

(S-PJ

(15.7)

Consider the amplitude and phase due to any pole or zero, for example,

* +jP, shown in Fig. 15.13. The magnitude is

and the phase

is

{a>,)

4>.{a>d

arctan

The amplitude and phase of the

(<<

- ft*}*

ft,

'~

(15.8)

"
l

overall function

(15.9)

is

M(wJ =

(15.10)

QIT^M)
and

<K>i)

= 2 tJfOi) - j ,,(>,)

(15.11)

452

Network

analysis and synthesis

where the subscripts p and

due to a pole and

z indicate the contribution

zero, respectively.

Input
1. The numerator E(s) and denominator F(s) can be read in either as
polynomials (high degree to low) or in terms of their roots. If numerator
and/or denominator are read in as polynomials, their roots will be printed.

If they are given in

terms of their roots, the program could generate

polynomials from these roots.

2.

The data points can be read in if unequally spaced;

only the

minimum

point and the spacing need be read

Read Nl, ND,

if equally

spaced,

in.

NN

KK, LI, L2, L3

Read CN,

KK =
Test

Read

L1

CD

KK

KK=1

Read WMIN,

(I)

=
Test LI

Read A (I), B (I)

Ll

DW

Read E(l)

J
L2

=
TestL2

Read C (I), D (I)

L2=l

Read

F(l)

*~r*
To main program

FIG.

15.14.

Flow chart of input

instructions for magnitude

and phase program.

Computer techniques

A flow chart listing the input instructions is given in

initions

453

in circuit analysis

Fig. 15.14.

Def-

of the symbols in the flow chart follow.

= number of frequency points

ND = degree of denominator
NN = degree of numerator
KK = indicates equally spaced data points. We

Nl

Wjqn, Aa>, and number of points.

indicates unequally spaced data points.

then read in only

We

then read in

all

to,.

LI

=0

Read
read

L2

=
=

Read
Read
read

in zeros (if

complex conjugate, both zeros must be

in).

in numerator polynomial, high degree to low.

in poles (if

complex conjugate, both poles must be

in).

= Read in denominator polynomial, high degree to low.

CN = numerator multiplier
CD = denominator multiplier
1

Output
The main output of

the program is the frequency response consisting

of the following columns: frequency in radians W(I); amplitude,
M(W(I)); magnitude in decibels, 20 log10 Af; and phase in degrees,
^(W(I)).
typical printout of an amplitude and phase program is given
in Fig. 1S.6, which shows the magnitude and phase of the band-pass filter

designed using Szentirmai's program.

15.3

A FORTRAN PROGRAM FOR THE ANALYSIS OF

LADDER NETWORKS

In this section we will discuss the methods and organization for a

Fortran computer program for the analysis of linear ladder networks.
The analysis proceeds by computing the voltage and current at the input
of successive

sections, beginning

with the terminating section.

branch impedances of the ladder by comimpedance

arms according to the instructions of the
C
individual programmer.
The poles and zeros and the frequency responses of the input impedance and voltage transfer ratio at the input of each L section are
found. In addition, the program provides for the analysis of short- and
open-circuited networks as well as for those with normal terminations.
Separate problems may be run consecutively if so desired.

The program

bining R, L,

calculates the

series


Network

454

vm

analysis and synthesis

Zm

v -l

Zn

la"

Vo

Vl
l

Zi

1
1

Zn-l
|

|
1

Ym

Yi

r-i

Y
l

FIG. 15.15

initially decomposed into separate L sections as in

L sections are then added successively to the terminating
form the complete network. With the addition of an L

The network
Fig. 15.15.

is

These

section to

and current
computed by the equations
section, the voltage

^n =

at the input of the resulting

network are

(15.12)

^n^n

+ Vn-1

which were originally discussed in Chapter 9 of this book. Thus the

program proceeds toward the front of the ladder requiring only the branch
immittances of the present L section and the voltage and current of the
previous

section to

make

its

calculations.

Suitable assumptions for the initial voltage

and current

V and

/ allow

for analysis of short- and open-circuited networks as well as for those

with normal terminations. These initial values of voltage and current are
determined by control instructions.
For the calculation of the voltage and current at an L section, the

branch impedances of the section are required. To simplify the preparation of the input data specifying these impedances, the following procedure
is used. Each branch impedance is formed by the addition in series or
parallel of basic R, L, C impedance arms. A basic R, L, C impedance
arm is a series combination of a resistance, an inductance, and a capacany or all of which may be absent. The impedance arms are speciitor
fied by the element values R, L, C and an instruction indicating how the
arm is to be combined.
The elements of the impedance arms are frequency and impedance
normalized to aid computation, and then the arms are combined as
specified to form the appropriate branch impedance.
The roots of the voltage and .current polynomials for each L section

are

computed by a root-finding

routine.

Computer techniques

455

in circuit analysis

For the normal load and open circuit termination, the frequency
responses of the input impedance, and voltage-transfer ratio are computed
for short-circuit termination, the input impedance and current gain are
computed. Provision is made for either a linear or logarithmic increment

The frequency boundaries and increments are read by the

program along with the input data.
According to instructions given by the programmer, the various calin frequency.

culations are printed for each

L section or only for the network as a whole.

Program operation
The ladder network-analysis program requires two sets of input data.
The first set consists of the control instructions. These tell the computer
which of the various options, such as information concerning polynomial
roots or frequency responses, are to be exercised.
In addition, the control instructions provide the computer with necessary
parameters such as normalizing factors, and the number of L sections in
the ladder. The second set of data determines the branch impedances of
the network by specifying the R, L, C impedance arms and their combining
instructions.

Impedance data
The data specifying

the branch impedances consist of an ordered series

an impedance arm and/or a comarm of any branch impedance requires no

of instruction cards, each determining

bining instruction.

The

first

combining instruction. The branch impedance is set equal to the impedance of this arm. Subsequent impedance arms are added in parallel
or series with the existing branch impedance, according to the combining
instructions of the arms. A blank card tells the computer that a complete
branch impedance has been formed and that the next arm begins a new
branch impedance. If blank cards are not properly inserted, the computer
will calculate

a single giant branch impedance.

the data determining the branch impedances isZx ljYlt

The order of
^a

1/

Jg,

The values of R, L, C and the combining instruction XIN are written

on one card in the order XIN, R, L, C.
The values of XIN and the associated operation are shown in the
table at the bottom of page 456.
Although the instructions XIN = 1.0, 2.0 will suffice for the conmany branch impedances, they are inadequate for other

struction of

impedances.
For example, suppose the series combination of two tank circuits (Fig.
15.16) is desired as a branch impedance, this impedance may only be

Network

456

anal/sis

and synthesis
50m*f

lMf

2mh

.05 h

-^OOOy-

Tank

circuit

Tank

circuit

XIN
5.00000

1.0

2.00000E

2.0
3.0

11

- 03

.05

.000001

2.0
4.0

Card

Blank

FIG. 15.16. Construction of branch impedances using instructions for XIN.

constructed with use of three

and an additional

Tank

circuit

more

instructions

(XIN

3.0, 4.0,

and

5.0)

of computer storage locations.

formed in the usual manner, but must then be tem-

set

A is

porarily stored during the calculation of tank circuit B.

The two tank

then added to form the final branch impedance. Examples of

impedances that cannot be formed are shown in Fig. 15.17. Observe that
circuits are

a series branch impedance of zero ohms often simplifies the formation of

branch impedances. Two blank cards will indicate a short-circuited
impedance.

XIN
Blank card

Operation

A complete
the next

branch impedance has been calculated, and

begins a new branch impedance.

arm

1.0

This arm is added in series with the existing branch

impedance.

2.0

This

3.0

This arm begins a new internal branch impedance.

Subsequent XIN = 1.0 and 2.0 refer to this new internal branch.

4.0

Add

arm is paralleled with the existing branch impedance.

the two internal branches in series to form the final

branch impedance.
5.0

Parallel the internal branches.

Computer techniques

HWV-i

in circuit analysis

457

r-VW-i

-Wrr-VArn

[-VS/Sr-j

L-vs/Sr-l

FIG.

15.17.

Branch impedances that cannot be formed by program.

Output
coefficients of the voltage and current polynomials are printed in
frequency normalized form. To calculate the denormalized coefficients,
divide by the normalizing frequency raised to the power of the corre-

The

sponding exponent.

and zeros of the voltage and current polynomials are in

frequency normalized form to facilitate zero-pole plots.

The

poles

Frequency responses appear in denormalized form. The frequency

variable is in radians per second. The impedance level is in decibels,
the phase in degrees. The gain and phase of the transfer ratios are
expressed similarly.

15.4

PROGRAMS THAT AID IN DARLINGTON

FILTER SYNTHESIS

Two double precision Fortran programs have been written that help
in the synthesis of double-terminated filters using the Darlington procedure. The mathematics of the programs is described here.
Given a forward transmission

coefficient

S8i(s)

KN(s)
D(s)

(15.13)

Network

458

we

analysis

and synthesis

generate an input reflection coefficient

Su(s) Sn(-s)

D(s)

Su(s) from

the equation

S(s) S gl (-5)

D(-s)

K* N(s) N(-s)

(15.14)

D(-s)

D(s)

The denominator of Su(j) is D(s) because all the poles of Su(s) must be
in the left-half plane. The zeros of Sn(s) are not so restricted. If Sn(s)
Su(s) has zeros &tajb and a jb, the zeros of Sn(s) can be chosen
as either a jb or a jb. The only difference is that certain choices
lead to niters with unity-coupled coils and others without. 12
Once the zeros of Sn(s) are chosen, the input impedance of the filter
is

then
Zi(s)

1
1

ReadCN

~
+

Sii(*)

(15.15)

Su(s)

Numerator

multiplier

I
Read

CD

Denominator

Number

ReadLN

I
Read

A (I),

B(I),N(I)

Read LD

multiplier

of zeros

(complex conjugate zeros

count as one zero)

Read

real

and imaginary

parts of zeros and

multiplicity,

N (I)

Number of poles
(conjugate poles
count as one pole)

Read
H(I),G(I), M(l)

Read

real

and imaginary

parts of poles and

multiplicity, P(l)

To program

FIG.

11

15.18. Input to

T. Fujisawa, "Realizability

Sn(5)

Theorem

Ladders without Mutual Induction," Trans.

1955, 320-325.

5u (j) program.
Low Pass
on Circuit Theory, CT-2, December

for Mid-Series or Mid-Shunt

IRE

Computer techniques

459

in circuit anal/sis

Program descriptions

We

next proceed with brief descriptions of the two programs. The

program finds the roots of Su (s) S^s) given the roots of N(s),
N(s), D(s), and D( s). A flow chart for the input of the program is
given in Fig. 15.18. Note that we must read in both the zeros of N(s) and
N(s), although only half of a complex conjugate pair need be read in.
The same applies for the roots of Z>($) and /)($). The constant CN is
first

K* in Eq. 15.14.
The second program performs

the computations in Eq. 15.15 for

combinations of zeros of Sxl(s), given the fact that the zeros of
Su(s) need not be in the left-hand plane. The previous program finds all
the zeros of Sn (s) Su ( s). We must choose only half the number of
zero pairs for 5u(s). Certain combinations of zeros of Sn (s) Su (s)
lead to filters with coupled coils; others do not. If one wishes to try a
number of combinations of zeros for 5u(f), the program has the facility
to enable him to do so. He need only supply those zeros of Sn(s) Su ( s)
in the right-hand plane, C(I) +y'B(I), and a list of constants S(I), which
different

are either 1.0 or

S(I)

1.0 so

that the zeros of

A(I) +y'B(I).

C(I) +y'B(I)

denominator polynomials of

The

Sxl(s) may

be represented as

routine forms numerator

Su (s)

= E(s)

(15.16)

F(s)

ReadLD

Read

ReadLN

Read

of pole pairs of

Su

count as one pair)

of

Sn

and

multiplicity,

M (I)

Number

of right-hand plane

zeros of

Sn (a) Sn(-s)

Real and imaginary parts of right-hand

C(l), G(l), N(l)

Number

(real poles

Real and imaginary parts of poles

H(I).G(I),M(I)

plane zeros and multiplicity,

N 0)

I
Read SO)

Multiplicative constants

(i 1.0) (each
combination of zeros has LN cards)

To program

FIG.

and

Su(s)

15.19.

Input to

Zta(j) program.

460

Network

analysis

and synthesis

and then computes the numerator and denominator of Z^s) as

ZM= i^ = m^m
The input

summarized on the flow chart

instructions are

(15 17)
.

in Fig. 15.19.

Problems
15.1
Organize a flow-chart for computing magnitude and phase given only
the unfactored numerator and denominator polynomials. Do not use a rootfinding subroutine.

15.2

Write a program to calculate the delay of

_,.

at the points o>

= 0,

1, 2,

3(5

1)

10.

Repeat Prob. 15.2 for a general transfer function given in terms of its
unfactored numerator and denominator polynomials. The frequency points co t
are to be read in by the computer.
15.4 Suppose you have calculated the phase of a transfer function at points
m = 0, 1, 2, .... 50. Devise an algorithm to test for phase linearity. The
deviation from phase linearity is to be called phase runoff.
15.5 Write a program to analyze a two-mesh network made up of only
15.3

R, L, and
15.6

C elements.

Repeat Prob. 15.5 for nodal analysis.

15.7 Write a program to calculate the step and impulse response of a linear
system whose system function contains only simple, real poles and zeros.

Repeat Prob. 15.7 if simple complex poles and zeros are allowed.
15.9 Write a program to calculate the residues of a transfer function with
multiple as well as simple poles. The poles and zeros are to be real.
15.8

appendix

Introduction to matrix algebra

A.I

FUNDAMENTAL OPERATIONS

Matrix notation

is

merely a shorthand method of algebraic symbolism

that enables one to carry out the algebraic operations

more

quickly.

The theory of

matrices originated primarily from the need (1) to solve

simultaneous linear equations and (2) to have a compact notation for

linear transformations

from one

As an example of (2),

of variables to another.

set

consider the set of simultaneous linear equations

u*i
^l^X

+
+

ami^i

aM";3

+
+

amixa

i**a

+
+

alnxn
"
a* *

= yx

= y*
**

(A.1)

+ amnxn = ym

These may express a general linear transformation from the xt to the yt

In general, mj&n. An example where the numbers of variables in the
two sets are unequal is that of representing a three-dimensional object in
two dimensions (in a perspective drawing). Here, m = 2 and n = 3.
.

Definition

an ordered rectangular array of numbers, generally the

The matrix is denned by giving
all its elements, and the location of each.
The matrix of the equations in Eq. A.1 written as
matrix

coefficients

is

of order

is

of a linear transformation.

mx

n.

(The

Oil

12

fl gl

att

first

ln

number here
461

is

the

number of rows; the

462

Network

second

is

the

analysis

and synthesis

number of columns.) The matrix may be denoted by a

A or by [aw ].

single capital letter

matrix

is

a single complete

entity, like

a position in chess.

Two

and B are equal only if all corresponding elements are the

same: atj = btj for all i andy. A matrix may consist of a single row or
single column. The complete matrix notation applied to Eq. A.1 is
matrices

<hi

olt

Oi

r*r

On.

on

oin

xa
Xn
i

'mi

'ml

"yr

=
.

Vt

(A.2)

Vm .

says, if put crudely, that A operates on x t to yield y,. This

emphasizes the similarity between a matrix and a transformation. The

which

and y-matrices are column matrices.

and column matrices are called vectors (specifically, row vectors
and column vectors) and their similarity to the more usual type of vector

x-

Row

is

Here, vectors will be written as small

discussed later.

letters,

such as

that the elements of vectors need only one subscript, while

or y. Note
elements of matrices need two.

A.2

ELEMENTARY CONCEPTS

Square matrix

square matrix has the same number of rows as columns (i.e., a

matrix of order rat). The Y matrix is an example of a square matrix
(A.3)

Diagonal matrix

diagonal matrix

diagonal are zero

matrix

is

is

(i.e.,

a square matrix whose elements off the main

for 1 ?*/) The following
one in which <xw =

diagonal.
'1

01

0-2

(A.4)
3

_0

Unit matrix

A unit matrix

a diagonal matrix for which a u

is
denoted as U. For example,
is

for

=j, and

is

0"

"1

U=

(A.5)

,0

1_

Appendix

463

Equality

Two matrices are equivalent if they

columns and
Suppose yx =

2,

in matrix form,

have the same number of rows and

the elements of corresponding orientation are equal

if

yt

3, and ya

6.

If we write this set

of equations

we have
"

2"

~Vi

= -3

y*

(A.6)

-6.
.y*.
Transpose
The transpose of a matrix A denoted as A r is the matrix formed by
interchanging the rows and columns of A. Thus, if we have
,

A=

-6

(A.7)

-CIS

A*

then

(A.8)

Determinant of a matrix
The determinant of a matrix

is defined only for square matrices and is

formed by taking the determinant of the elements of the matrix. For
example, we have
1
2

12

det

l
4j

L-5

14

(A.9)

Note that the determinant of a matrix has a particular

itself is merely an array of quantities.

value, whereas the

matrix

Cofactor

The

cofactor

deleting the fth

A i}

of a square matrix

is

the determinant formed by

row andy'th column, and multiplying by (l)i+i

example, the cofactor

A tl

-n
An =

IS

For

of the matrix

(-iy*+i

6= -6

(A.10)

(A.11)

Adjoint matrix

The

A is formed by replacing each

and transposing. For example, for the

adjoint matrix of a square matrix

element of

by

its

cofactor

Network

464

and synthesis

analysis

matrix in Eq. A.10,

we have

adjA

h
i

l:

-l;

(A.12)

Singular and nonsingular matrices

A singular matrix is a square matrix A for which det
singular matrix is one for which det A ?* 0.

OPERATIONS

A.3

ON

A=

0.

non-

MATRICES

Addition

may be added

both matrices are of the same order.

added to the element of the second
Each element of the first
orientation
is the same. An example of
column
and
matrix, whose row
A.
13.
shown
in
Eq.
is
addition
matrix

Two

matrices

if

matrix

is

(A.13)

[-in-'-H-<-2
Thus

if

matrices A, B,

A+B+
The

associative

K are all

+K=

of the same order, then

+b +

[a u

it

+k

tj ]

(A.14)

and commutative laws apply.

Associative:

Commutative:

A + (B + C) = (A +
A+B=B+A

B)

+C
(A.15)

Multiplication by a scalar

We define multiplication by a scalar as

XA

Thus to multiply a matrix by a

A[a w ]

[AaJ

scalar, multiply

(A.16)

each of

its

elements by

the scalar.

Example

Al
"

-1

"

0"
1

-3 2

0"

-3

-9

(A.17)

Appendix

465

Linear combination of matrices

of addition and multiplication by a scalar are combined,
for two matrices of the same order we have
If the rules

oA
where and

/SB

[*ait

+ pb

(A.18)

it ]

are scalars.

/9

Multiplication
In order for matrix multiplication AB to be possible, the number of
columns of the first matrix A must equal the number of rows of the second
matrix B. The product C will have the number of rows of the first and
A has
the number of columns of the second matrix. In other words, if
product
the
then
columns,
rows and n columns, and B has n rows and/>

m
C

will

have

rows and

columns.

The

individual elements of

are

given by

(A.19)

Example A.2
4
(A.20)

2
[-:-:;]-[-,:;]
-3

Example A3.

The system of equations

u/i

ZtJi

+ Ws = Vx
+ z/s = Vt

(A.21)

can be written in matrix notation as

(A.22)

caca-ra
We

see that systems of equations

can be very conveniently written in

matrix notation.

Matrix multiplication

is

not generally commutative, that

AmnBB * B^A,
Observe that the product BA is not defined unless p =
of square matrices

is

generally not commutative, as

is,

(A.23)

m. Even a product

may be

seen in the

following example.

L-i

onr-i
2JL

<n

_ r-i

2j~L

oi
(A.24)

f]

Network

466
If

we

anal/sis

and synthesis

interchange the order of multiplication,

1

-1

-1

-2

-1
2

we

obtain

(A.25)

Because of the noncommutative nature of matrix multiplication, we

must distinguish between premultiplication and postmultiplication. In
BA, A is premultiplied by B; B is postmultiplied by A. For matrix
multiplication, the associative and distributive laws apply.

= ABC
+ C) = AB + AC

Associative:

A(BC)

Distributive:

A(B

(AB)C

(A.26)

Transpose of a product
The transpose of a product AB is equal to the product, in reverse
of the transpose of the individual matrices A and B, that is,
(AB)

(ABC)r

and

- B rA T

order,

(A.27)

= C T(AB) T = C^B^A 2

"

(A.28)

The product x Tx, if x is a column vector, is a scalar number equal to

the sum of the squares of the elements of x. Thus we have

xT x

The product xx r

Common
(a)

[ Xl

*i

xt

+ x*
T
is a square matrix C such that C = C
+

*,*

It is also

xx r

if

is

expression auxf

where

a and b are column vectors.

(b) The sum of squares x^

The

aA + ajb +

typical element of a product matrix,

(c)

expressions in simple notation

The sum of products

vector).

(A.29)

x]

may be

+ xt* +

+ anb,

written as

x n*

a nnxn *

is

which are the

a^b or b ra where

x Tx

(x,

column

x rAx

(x,

column

thus

a row vector.

+ atixta +

is

a diagonal matrix with elements ait Similarly, the

T
T
expression aaxjfx
a^e^yt H
h a nnxyn is x Ay or y Ax.
(d) An expression such as
vector),

is

2 Z aux x
i

t>

a,

Appendix
is

a quadratic form. This

ii*i*

467

is

1a x%xx xt

*!*

+ 2a ln x1 x

2a u xt xa H

<J*

h 2a, B a;2 x B

(A.30)

x TAx

Inverse
Division

is

not defined in matrix algebra. The analogous operation

that of obtaining the inverse of a square matrix.

matrix

obtain

= AA- = U
1

A-1 we first obtain the adjoint

,

the determinant of A.
that

inverse

A-1

is

of a

A is defined by the relation

A-*A

To

The

The

inverse

A-1

is

(A.31)

of A, adj A. Then we obtain

equal to adj A divided by |A|,

is,

A" 1

adj

(A.32)

|A|

Example A.4.

Let

A be given as
(A.33)

Its

determinant

is

|A|

=3

(A.34)

and the cofactors are

An = 1
Afn = 1
The

adjoint matrix

Au =

An =

(A.35)

is

adJA =

[-l

2j
(A.36)

-C
so that

A-1 is

-C 1

-EH

(A.37)

Network

46.8

As a check we

and synthesis

anal/sis

see that

-r

"

'

-1

-l

(A.38)

_i

If the determinant

of the matrix is zero, then the inverse is not defined.

other words, only nonsingular square matrices have inverses.

In

SOLUTIONS OF LINEAR EQUATIONS

A.4

Consider a

set

of linear algebraic equations, to be solved simultaneously.

011*1
asi^i

m*i

The h t are
we have

constants.

H
+
+ a^xt + +
r-

<*iss*

+ a n **i +

gives

In expanded form, this

In matrix notation

(A.40)

=A

(A.41)

Ax
Premultiplying by

(A.39)

+ a nnxn = h n

desired to solve for the xf

It is

-1

= hx
aSnx = hs
alnxn

is

x.

^si

J_

ia

|A|

^32

-"m

(A.42)

Thus
|A|

*1

Oia

*is

ht

On

ha

a S2

|A|

and

similarly for

xt and *3

This

is

the familiar Cramer's rule for solving

such equations.

Example A.5.

Solve for x, y,

z.

2x

+2 =2
+y =

3y

+ = -1

-* +

(A.43)

Appendix
In matrix form, these equations are written as
1

-1

-1
Also, |A| = 10. Thus,

-&

Therefore

A.5

The following

is

A_1h
-r

7-2

-i

-0.4, z

(A.44)

-l

= 0.7, y =

REFERENCES

-2

469

Ax = h where

~x~

we have x =

*1
y

r
=

X
io

-4

(A.45)

= 0.9

ON MATRIX ALGEBRA

a short

list

of books on matrices that the reader might wish to

examine.
A. C. Aitken, Determinants and Matrices, 9th Ed., Interscience Publishers,

New

York,

1956.

R. Bellman, Introduction to Matrix Analysis, McGraw-Hill Book Company, New

York, 1960.
York,
R. L. Eisenman, Matrix Vector Analysis, McGraw-Hill Book Company, New
1963.

D. K. Faddeev and V. N. Faddeeva, Computational Methods of Linear Algebra, W. H.

Freeman and Company, San Francisco, 1963.
Publishers,
F. R. Gantmacher, Applications of the Theory of Matrices, Interscience

New

York, 1959.

Hohn, Elementary Matrix Algebra, The Macmillan Company, New York, 1964.
McGraw-Hill Book
L. P. Huelsman, Circuits, Matrices, and Linear Vector Spaces,
Company, New York, 1963.
and Sons, New York,
P. LeCorbeiller, Matrix Analysis of Electric Networks, John Wiley

F. E.

1950.

Survey of Matrix Theory and Matrix Inequalities, Allyn and

Bacon, Boston, 1964.
Sons, New York,
E. D. Nering, Linear Algebra and Matrix Theory, John Wiley and

M. Marcus and H. Mine,

1964.
1952.
of Matrices, Addison-Westey, Reading, Massachusetts,
Englewood Cliffs, NJ.,
L. A. Pipes, Matrix Methods for Engineering, Prentice-Hall,

S. Perlis, Theory

1963.

A. M. Trapper, Matrix Theory for

Electrical Engineering Students, Harrop,

London,

1962.

A. von Weiss, Matrix Analysis for

NJ., 1964.

Electrical Engineers,

D. Van Nostrand, Princeton,

appendix

Generalized functions and

the unit impulse

GENERALIZED FUNCTIONS

B.I

The

unit impulse, or delta function,

is

a mathematical anomaly. P. A.

M. Dirac, the physicist, first used it in his writings on quantum mechanics. 1

He denned

the delta function d(x)

d(x)
Its

most important property

by the equations

d(x)

dx

(B.I)

for

jt

is

f(x)6(z)dx=f(0)

(B.2)

is continuous at x
0. Dirac called the delta function an
improper function, because there existed no rigorous mathematical justi-

where f(x)

it at the time. In 1950 Laurent Schwartz* published a treatise

The Theory oj Distributions, which provided, among other things,
a fully rigorous and satisfactory basis for the delta function. Distribution
theory, however, proved too abstract for applied mathematics and

fication for
entitled

P.

A. M. Dirac, The Principles of Quantum Mechanics, Oxford University Press,

1930.
1

and

L. Schwartz, Theorie des Distributions, Vols.

1951.

470

and

II,

Hermann

et Cie, Paris, 1950

Appendix B
physicists.

It

was not

until 1953,

when George Temple produced a more

elementary (although no

less rigorous)

ized functions,* that this

new branch of

Our treatment of

deserved.

theory through the use of generalanalysis received the attention

it

generalized functions will be limited to the

definition of the generalized step function

The treatment of

471

and

its

derivative, the unit

work of
Temple4 and Lighthill. 8
To get an idea of what a generalized function is, it is convenient to use
as an analogy the notion of an irrational number a beng a sequence {oc}
of rational numbers <x n such that
impulse.

these functions follows closely the

= lim a
n-ao

where the

limit indicates that the points

the point representing a.

<x

on

the real line converge to

on the
on the sequence {a} defining
a.' We can also think of a generalized function as being a sequence of
functions, which when multiplied by a test function and integrated over

number a

irrational

(00,

are actually performed

a finite limit. Before we formally define a generalized

important to consider the definition of (1) a testing function

oo) yields

function,

and

All arithmetic operations performed

(2)

it is

a regular sequence.

DEFINITION

A function #f) of class

B.I

everywhere, any

is differentiable

any of its derivatives are multiplied by

lim

[t

m
<f>

C[#f) G C]

number of times and

M(t)]-*0

raised to

for all

is

one that

that (2)

when

any power, the

m&fe^O

it

(1)

or

limit is

(B.3)

*->oo

Any

testing function is

a function of class C.

The Gaussian function e-*'ln% is a function of class C. It is

obvious that if a function is of class C then all of its derivatives belong to class C.
Example B.l.

*G. Temple, "Theories and Applications of Generalized Functions," J. London

Math. Soc., 28, 1953, 134-148.
4
G. Temple, "The Theory of Generalized Functions," Proc. Royal Society, A, 228,
1955, 175-190.
6

M.

and Generalized Functions, Cambridge University

book to "Paul Dirac, who saw it must be
Laurent Schwartz, who proved it, and George Temple, who showed how simple
J. Lighthill,

Press, 1935.
true,
it

Fourier Analysis

Lighthill dedicated his excellent

could be made."
* This defines an irrational

number according to the Cantor definition. For a more

on real variables such as E. W. Hobson, The Theory of
Functions of a Real Variable, Vol. I, third edition, Chapter 1, Cambridge University
detailed account see

Press,

any

text

Cambridge, England, 1927.

Network

472

and synthesis

anal/sis

DEFINITION

A sequence {/(*)}

B.2

to be regular if for any function

of functions of class

is

said

belonging to C, the limit

<f>(t)

lim(/,#=lim
n-*ao

J oo

n-*oo

fn(t)<Kt)dt

(B.4)

Note that it is not necessary that the sequence converge pointwise.

For example, the sequence {e~ nt\nlir)^} approaches infinity as n - oo at
the point t = 0. However, the limit lim (/", <f>) exists.

exists.

n-oo

DEFINITION
alent if for all

B.3

Two

regular sequences {/} and {g} are equiv-

lim(/B ,#=lim(g B ,#
n-*oo

Example B.2.

regular sequences {e-n**(/f) w }

The

(B.5)

n-*ao

{<r' ,/aB*(l/V2^i)} are

and

equivalent.

DEFINITION B.4
generalized Junction g is denned as a total, or
complete, class of equivalent regular sequences. The term total implies
here that there exists

no other equivalent

Any member
both g and the

to this class.
represent

denning g.

We denote this

Example B.3.

total class

example, {g n },

is sufficient

to

of equivalent regular sequences

symbolically by the form

1 '"4

Or*

},

~ {gn

}.

function

~ {e~**

The inner product

B.5
<f>(t)

{--/*}]

{e-"'-}

same generalized function g

DEFINITION
g and a

regular sequence not belonging

class, for

All of the equivalent, regular sequences

[{*-'*/},

represent the

of the

C is defined

(g,

lnt
}.

<f>)

of a generalized function,

as

(g,#=lim f"gn(t)<Kt)dt
n-*oo

The inner product

is

(B.6)

J co

often given the following symbolic representation.

(g,

Note that the

+)

(B.7)

g(t)<Kt)dt

integral here is used symbolically

and does not imply

actual integration.

DEFINITION

g and h are two

g
{g n } and h
representation g + h
{g +
B.6

If

resented by the sequences

defined by the

Note

that the set of sequences {g n

equivalent regular sequences

and

h; therefore

+ h n}

generalized functions rep-

~ {h

n },

the

sum g

+h

is

A}.

represents a total class of

made up of the sum of sequences

defining

+ h is a properly defined generalized function.

DEFINITION
and a constant a

B.7
is

The product a.g of a

is

473

generalized function;

~ {#}

defined by the representation ag <~ {gn }.

DEFINITION
{#}

Appendix B

g~

B.8 The derivative g' of a generalized function

{g' n }.
defined by the representation g'

Example B.4. For the generalized function

by

gx ~ {*-*

1/b *}

the derivative

is

represented

and

(g\, *)

= lim

"
(

- ^ e"**'"1

*(/)

(B.8)

//

we have defined the operations of addition,

and differentiation. It must be pointed out that
the operation of multiplication between two generalized functions is not
In Definitions B.6, B.7, and B.8
multiplication

by a

scalar

defined in general.

We

next consider an important theorem, whose proof

Lighthill, 7

which

as a step function

Theorem B.l.

by a generalized function

for

C.

lent in terms

write

satisfying the condition

dt

<

co

(B.9)

N ^ 0, there exists a generalized function8 / ~ {/,(/)} such that

some

for all

1/(01

given in

equivalent.

Given any ordinary function /(f)

1'

is

any ordinary function, such

will enable us to represent

/ =/.

In other words, an ordinary function satisfying Eq. B.9 is equivaof inner products to a generalized function. Symbolically, we
If, in addition, / is continuous in an interval, then lim fn

=/

""*"

pointwise in that interval.

Furthermore,
multiplication,

it

and

can be shown that all the operations of addition, scalar

differentiation performed on both /and /yield equivalent

results, that is,

(a

when

differentiation is permitted

m'

= (/i +

6ti'

on the ordinary

<M

function.

'

Lighthill, op. cit., Section 2.3.

Note that when we represent an ordinary function by generalized function equivalent,

we use a bold face

italic letter

to denote the generalized function.

Network

474

DEFINITION
total class

analysis

B.9

and synthesis

The

generalized step function u

of equivalent regular sequences

(,#

= lim
n-*oo

{(*)}

is

defined as the

such that

fuJit)4(t)dt
J ao

-f"ii(t)#0A

(B.12)

J 00

where

u(i) is the unit step defined in

Chapter

That

2.

{(/)} exists is

guaranteed by the previous theorem allowing representations of ordinary

u.
functions by generalized functions. Hence, we write u

Example B.5.

The sequence

/>o
t0
which

is

plotted in Fig. B.l,

is

one member of the

class

(B.13)

of equivalent regular

sequences which represents the generalized step function.

FIG. B.l. The generalized step sequence,

DEFINITION
is

,(/)

B.I0 The unit impulse, or Dirac delta function <$(r),

{'(*)}.
denned as the derivative of the generalized step function d(t)

Appendix B
It

475

should be stressed that <J(f) is merely the symbolic representation for a

of equivalent regular sequences represented by {u' (t)}. Thus
n

total class

when we

we

write the integral

actually

mmdt

mean
f " <5(0 <Kt) dt

= (d,

J oo
Example B.6.

<j>)

= lim
n-oo

f " u' n(0 #t) dt

The sequence

^-(5-2r)p[-i(J

,>o

+ |.)]

(B.15)

=
in Fig. B.2

is

(B.14)

J oo

one member of the class of equivalent regular sequences which

Other members of the class are the sequences

represents the unit impulse.

{e-*(MK} and {^1*^(1/ VJ^n)}.

1.0

1.2

1.4

1.6

_1_

_L_

FIG. B.2. The generalized sequence,

'(/).

476

B.2

Network

anal/sis

and synthesis

PROPERTIES OF THE UNIT IMPULSE

Sifting

The most important property of the

unit impulse

is

the sifting property

represented symbolically by

J<o

W(0*-/(0);

differentiable over [a,

where /is any function

Eq. B.16

is

(B.16)

(M,lj8|<>)

The

/J].

left

hand

side of

defined formally by

f>
\ \t)f(t)dt
Js<0

slim

the sifting property

The proof of

n-ao
is

(B.17)

\"*u'JLt)f{t)dt
Ja<0

obtained by simply integrating by

parts, as follows.

>O

lim
n<x>

r u'(0/(0 dt = lim u n(t)f(t)\- lim

J<0
n-oo

n-oo

f'ii,(0/'(0 dt
J*

-/tf)-f'limu.(0/'(OA

_,

= /C8)-[/(/5)-/(0)]=/(0)
Pictorially

impulse

is

we

represent <$(0

centered at

by a spike as shown in

Fig. B.3.

If the

= a, then the sifting property is given symbolically

as

d(.t

fl

)/(f) dt

- /(a)

(|a|,

KO

FIG. B.3. The unit impulse.

HI

<

ao)

(B.19)

Appendix B
where/'(0 most exist over
are

infinite,

we

actually

[a,

Note that when the limits of integration

/?].

mean

(S(t)f(t)dts lira \\t)f{t)dt

J oo
a-* oo Ja
In the

477

sifting property, if

both

a,

(I

>

or a,

dt

<K0/(0

/?

(B.20)

< 0, then

(B.21)

/:

The proof of this property is similar to the original proof of

property, and will be left as an exercise for the reader.

the sifting

Integration

The

defining equations of the delta function according to Dirac are

f>0
d(x)

dx

/.<o

(B.22)

d(x) = 0;

i*

These are actually properties of the delta function as viewed from the
generalized function standpoint. The proof can be obtained directly from
the sifting property. Suppose we have the integral
f>

1
and we

let f{t)

1.

/?

(B.23)

Then we have

Ia<0
If both a,

\t)f(t)dt=f(0)

d(t)dt=f(p)

(B.24)

are greater than zero or both are less than zero, then

1
This property

is

d(t) dt

stated symbolically

(B.25)

by the conditions

d(t)

for

ft 0.

Differentiation across a discontinuity

Consider the function /(0 in Fig. B.4. We see that /(f) has a discontinuity of A at t
T. If we let/i(0 =f(f) for t
T, and/i(f) =/(*)
A
for t
T, then we have

<

At)=MQ + Au(t-T)

(B.26)

Network

478

analysis

and synthesis

*M
h(t)

FIG. B.4. Function with discontinuity.

Since /(0,/i(0 and u(t) satisfy the condition

lg(OI

dt <
&\N
I (1 + O

for

some N; we can

co

represent these ordinary functions by generalized

functions
(ft)

Taking derivatives on both

=Mt) +

n(0

(B.27)

sides of Eq. B.27 yields

fV)=f\(.t)

(B.28)

Au'(.t)

which symbolically can be written as

no =A(o + A m
We

thus see that whenever

we

(B.29)

differentiate across

a discontinuity, we

obtain a delta function times the height of the discontinuity.

Example B.7.

The

step response of

Hf)

shown

in Fig. B.Sa.

an R-C network

is

= Ae-" T u(t)

The impulse response

is

shown

in Fig. B.Sb.

(B.30)

is

h'(t)=Ad(t)-j,e- l T u(t)
and

given as

(B.31)

Appendix B

479

FIC. B.5. Differentiation across a discontinuity.

Differentiation

The

derivative of

a delta function, which we

symbolically as d\t)
f"(i) exists over [a,

{u" n(t)}.

It

call

a doublet,

is

defined

has the following property, where

fit].

/V>o
<J'(0/(0 dt
J
J<0 '

The proof is obtained through

-/'(0);

(||,

\fi\

successive integration

<

oo)

by

parts.

(B.32)

>0

lim
T" V(0/(0 dt = n-oo
f "B(0/(0 dt
Ja<0
Ja<0
fi>0

- Km
We see that since

'.(0/(0

lim u'Jfi)

lim

-lim
<0

n-oo

u' n (a)

>0

limH' B (0/0)'
n-*oo

<p

(B.33)

Ct>9
u'n (t)f'(t)dt
J<0

= 0,

=0

(B.34)

480

Network

analysis and synthesis

f(t)

FIG. B.6. The doublet

<$'(')

We then integrate by parts again so that

>0

-lim

u' B

(0/'(0^=-lim n/'| +lim

['u n (t)f\t)dt
(B.35)

\yu{t)f{t)dt

-/'(/?)+

-fW +fW -/'(0) =

-/'(0)

In general, the derivative-sifting property can be stated symbolically as

*
-(*-)/()

J
J<0

where/ <B+1) (0 exists over [a,

The generalized function

A -(-DT"'()

(B.36)

/3].

d'(t)

is

pictorial representation of a doublet

sometimes called a doublet.

is

The

given in Fig. B.6.

Other properties of the

unit impulse
Dirac and others have obtained a host of identities concerning the unit
impulse. We will merely give these here without proof.

- ,5(0
d'(-t) - -<5'(0
t <3(0 = o
td'(t) = -d{t)
<5(-o

-a*) = i \a\-1 {d(t f(t)6(t-a)=f(a)d(t-a)

d(t*

(3)
(4)

(5)

a)

d(t

a)}

(6)

(7)

these properties are obtained through the inner product

with a testing function

(2)

diat)

The proofs of

(1)

The condition on/'" +1

<f>(t)

'

C.

is sufficient,

but not necessary.

appendix

Elements of complex variables

AND OPERATIONS

CI ELEMENTARY DEFINITIONS

A complex variable z is a pair of real variables (x, y) written as

z = x+jy

(CI)

where j can be thought of as v 1.

The variable x is called the real part of z, and y is the imaginary part of z.
Written in simpler notation, we have

The

Re(z),

variable z can be plotted

= Im(z)

(C.2)

on a pair of rectangular coordinates. The

abscissa represents the x or real axis, and the ordinate represents the y or
imaginary axis. The plane upon which x and y are plotted is called the

complex plane. Any point on the complex plane, such as z = 3 +j2, can
be represented in terms of its real and imaginary parts, as shown in Fig. C. 1
From the origin of the complex plane, let us draw a vector to any point z.
The distance from the origin to z is given by
|x|

and

is

known

(x*

y)*

(C.3)

as the modulus of

angle which the vector subtends

as the argument of z or

arg z

tan

-i*

V
x

is

The

z.

known
3 *

(C.4)

\z\,
we can
arg z and r
Letting 6
represent z in polar coordinates as

re

(C.5)
481

FIG.

CI

Network

482

Expanding

anal/sis

this last

and synthesis
obtain

+jr sin 6,

z == r cos 6

(C.6)

= r cos 8
y = r sin
x

so that

The

we

equation by Euler's formula,

two complex numbers

rule for addition for

(a

where j* = 1.

given as
(C.8)

+ jb)(c +jd) = (ac - bd) + j(ad + be)

(C.9)

If

we

+jd)

(c

When two complex numbers

(a

is

+ d)

+jb)

(C.7)

(a

are multiplied,

+j(b

c)

we have

express the complex numbers in polar form,

we

obtain

m
(c+jd) = r^ m
+ jb) =

(a

and

When we

we

(C.ll)

multiply the two numbers in polar form, then

r^rtf"*
If

(CIO)

ri e

divide these

two numbers

jv^**-

-*'

(C.12)

in polar form, then

7^ = 7^'^

(C13)

In rectangular coordinates, the operation of division can be expressed as

a+jb _ (a+ jb)(c - jd)

c+jd (c+jd)(c-jd)
ac + bd
.be ad
J
c + d*
c* + d*
In connection with the modulus of a complex number,

(C.14)

it is

useful to note

the following rules:

li2l

1*1*1*1

z-z*
where z*

is

= Uil
=
= |z|
|il

tal

l*i*l

the complex conjugate of z and

z*

is

2
l*il

(CIS)

defined as

= x +jy = x jy

(C.16)

Appendix

483

The following rules deal with operations involving the conjugate definition
Zi

+ zt = zx * + z,*

^=
Finally, if z has

*!**.*

(C.17)

a modulus of unity, then

z

-1

(C.18)

z*

The operations of raising a complex number to the nth power, or taking

the nth root of a complex number, can be dealt with most readily by using
n
the polar form of the number. Thus, we have z B
(re1")"
r eSnB ,

and

= jMne sn+tk,)M

k0l

(C

19)

ANALYSIS

C.2

If to each z
is

1/n

= x +jy, we assign a complex number w = u +jv, then w

a function of z or

w =/(*)
The following are examples of complex

(C.20)
functions,

i.e.,

functions of a

complex variable:

= 2z
w = log, z
w = 1/z
w = za + 4
w=
w

(C.21)

|z|

We

see that

w may

be complex, pure

real,

or pure imaginary, depending

upon the particular relationship with z. In general, the real and imaginary
parts of w are both functions of x and y. That is, if we let w = j + jv,
then

u-A*,V)
As an example,

let

we

v=f(x,y)

us find u and v for the function

w=
Simplifying,

and

z*

+4=

(x

+jy)*

+4

(C.22)

= z* + 4.
(C.23)

obtain

= (* - y* +/'2ry) + 4

(C^4)

Network

484

and synthesis

analysis

jy

jy

Ax

J&
z

+ Az

Pathl

+ A?

Path 2

-jy

-jy

FIG. C.2

so that

The

derivative of a

FIG. C.3
a:*

yl +

and

complex function /(z)

= 2y

i>

is

(C.25)

defined as

(C.26)
If

one

restricts the direction

we have what
function

is

is

known

or path along which

Az approaches

zero, then

as a directional derivative. However, if a

to possess a derivative at

all,

complex

the derivative must be the

same

any point regardless of the direction in which Az approaches zero. In

other words, in order for/(z) to be differentiable at z = z we must have

at

= constant

(C.27)

dz
for all directions of approach of Az.

Consider the two directions

and C.3. For path 1, we have
/'(z)

in

=lim

which Az approaches zero in

,/(z
lim'

A*->0 Av->0

If

we

+ Az)-/(z)

Figs.

C.2

(C.28)

Az

substitute

Az
into Eq. C.28,
f'(z)

we

= Ax + j Ay

(C.29)

obtain

+ A* +j(y + Ay)] -f(x+jy)

As + j Ay
/(z) = u +jv
f(z + Az) = u + Aw +j(v + A)

= lim

lim

f[x

(C.30)

Aac-0 Aff-*0

Since

and

(C.31)
(C.32)

Appendix

we

485

finally arrive at

/'()

Am +jAi>

= lim

lim

a-.o Av->o Aa;

+ j'Ay

=+

-*/Ao
= hm Au +
Ax
a*-o
,.

For path

2,

9u

(C.33)
.

dx

dx

we have

Au + Ad
Km " T J A
/

/'(z)=lim

a-oas-oAx

==lim

+ J Ay

AiLM

(C.34)

/'Ay

a-o
dt>

dy

du

dy

we assume that the function /(z) is differentiable, the

must be independent of path. Thus, we have
Since

_ =
dy

dy

From this

last equation,

we

4-

derivatives

(C 35)

9x

dx

obtain the Cauchy-Riemann equations, which

are

dv

du

dx

dy

du__dv
dx

dy

We have just seen that in order for a function to have a derivative, the
Cauchy-Riemann equations must hold. A function which is single valued
and possesses a unique derivative is called an analytic function. A set of
Cauchy-Riemann equations
For example, consider the function

sufficient conditions for analyticity is that the

are obeyed.

/(z)
/(z)

is

= z* + 4

(C.37)

analytic because

^ = 2x = ^
dy
9

dx

= 2

dy

(C.38)

dx

Network

486

On the

anal/sis and synthesis

= z* is not analytic because

other hand,/(z)

= +1
du

j.,

and

a
and

*>

SINGULARITIES

AND

(C39)

= -1

ay

ox
C.3

= y
1

RESIDUES

If/(z) is analytic within a region or domain in the

complex plane except

then/(z) has an isolated singularity at z
Suppose /(z) has a
singularity at z , then we can expand /(z) about z in a Laurent series

at a point z

- z )B

(z

-z
+

Om(2

- So)"* +

(C.40)

In the expansion,

term a_i

is

if

m is finite,

then z

is

called a pole of order

m.

The

called the residue of the singularity.

Example C.l.

Consider the Laurent series for the function /(z)

We can expand e* in a power series to give

= e*\z

about

the pole at the origin.

*=

I/i
z\

+* +

+ -*+

2!

3!

/
(C.41)

= l+l + Lz+l.z* + -z

According to the

2!

3!

Example C.2. Expand the function /(z)

and find the residue of the pole at z = 1
1

z(z-l)

(z

Note

then *o

is

|z

1|

that if

an

<

,
(z

1.

l/z(z

l)2

equal to

1.

about the pole at

1,

[1

1)

- (z -

1)

+ (z -

1)

- (z -

I)*

1)

is

equal to 1.

(C.42)

1)
1

<

is

- i)M + (z 1

for

of the pole at z

definition, the residue

,
1)*

- (z -

1)

(z

Here, the residue of the pole at z

we have an

infinite

essential singularity.

number of nonzero terms with

negative exponents,

Appendix
Example C3.

Find the residues of the poles at s

and s

C
-1

487
of the

function
s

To find

we simply perform a

the residues,

(C.43)

partial fraction

expansion

(C.44)

Thus the residue of the pole at

s = -lis +3.
C.4

+2

s =

is

3, and the residue of the pole at

CONTOUR INTEGRATION

In complex integration the integral is taken over a piecewise smooth path

C and is defined as the limit of an infinite summation

IC

f(z) dz

= lim 2 f(z
n-oo j 1

i)

A*/

(C.45)

where zt lies on C. Unlike the process of differentiation, the path along

which we take the integral makes a difference as to the ultimate value of
the integral.

Thus the

integral

r f(z)dz

(C.46)

upon whether we choose to inteshown in Fig. C.4. If we integrate along

in general, has different values depending

or path C8 as
a closed path, say from a to b and then to a again, we are integrating along
a closed contour. The path shown in Fig. C.5 is an example of a closed
contour. The following theorem, known as Couch/ s residue theorem gives

grate along path

a method for rapid evaluation of integrals on closed paths.

Jy

jy
Closed

'contour

FIG. C.4

Theorem C.l.

If

FIG. C.5

C is a simple closed curve in a domain

analytic except for isolated singularities at

z^, z,,

...

z,

D, within which /(z) is

then the integral along

Network

488

the closed path

anal/sis

C is

I
where

and synthesis

f(z)dz

= 2n/(JKi +K. + ---+KJ

(C.47)

represents the residue of the singularity z ( .

Example C.4.

Consider the integral

s+2 ds
sH.s + 1?
along the
within the

circle

\s\

(C.48)

= 2, as given in Fig. C.6. Since there are two singularities

= and at s = 1, whose residues are respectively 3

circle, at s

and +3, then the integral along the circle

S

s\s

is

+ 2 ds - 2*j(-3 +
+ iy

3)

(C.49)

C=|.|-2

Example C.5.
C.7.

The

Find the

function /()

integral of f(s) along the closed contour

is

35+5
IX* +

'*"-(* +

in Fig.

(C.50)
2)

Jy

'

'
L

shown

given as

-2 -1

-Jy

FIG .

C7

Appendix

489

A partial fraction expansion of f(s) shows that

m ri
+
+ + Txi
1

so that the residues are

and 2. The value of the

path within which both the

singularities lie is

j> /(s) ds
If f(z) is analytic in

along any closed path

iTTJil

a domain with no
is

zero, that

is

known

2)

integral along the closed

then

= 6tt/

singularities,

(C.52)

then the integral

is,

f(z)dz

if
This result

(C51)

as Cauchy's integral theorem.

(C.53)

D
Proofs of some theorems on
appendix

positive real functions

Theorem D.l.

If Z(s)

and W(s) are both

positive real, then Z(fV(s))

is

also

positive real.

When Re s 0, both Re Z(s) and Re W(s) ^ 0, then Re Z( W(s)) 0,

When s is real, both Z(s) and ff%s) are real, hence Z(W(s)) is real. Since

Proof.
also.

Z(W(s))

satisfies

Theorem D.2.

If Z{s) is positive real,

W(s)

Proof.

Theorem D.3.
Z(s)

Proof.

both conditions of positive realness,

1/* is positive real,

it is

positive real.

then Z(\/s) is positive real.

hence Z(tV(s)} = Z(l/s) is positive

real.

If lV(s) is positive real, then l/fV(s) is also positive real.

l/s is positive real,

hence Z(W(s))

l/fV(s) is positive real

by

Theorem D.l.
Theorem D.4. The sum of positive real functions is positive real.
Proof. Suppose Zx(s) and Z2(s) are both positive real. When Re s

ReZx k

then

ReZ2 ^

ReZj + ReZa = ReZ ^

so that

is

and

^ 0,

0.

Also, when s is real, both Zx and Z2 are real. The sum of two real numbers
a real number. Therefore, Zx + Zt is positive real.

Theorem D.5.
(i.e., lie

poles

and zeros of Z(s) cannot have

positive real parts

the right-half plane.

is a pole s
expansion about s so that

Suppose there

Proof
Laurent

The

in the right half of the s plane).

series

(*

- 5o)n

(S

- * )"

490

Let us

make a

Appendix

491

ReZ<)

FIG. D.I

where n is real and

approximated by

finite.

In the neighborhood of the pole S& Z(s) can be

Z(s)

,-^V(s - s n
)

We can represent Z(s) in polar form by substituting each term by its polar form;
(j - o)+" = rV and k_n = Kt** so that

i.e., let

z<*) = ^ **~**.

Re Z{s) =

* cos (* -

nfl)

which is represented in Fig. D. 1 When 9 varies from to lit, the sign of Re Z(j)
change 2n times. Since ReZ(s) t.
when Re s 2. 0, it is seen that any
change of sign of Re Z(s) in the right-half plane will show that the function is not
positive real. Therefore, we cannot have a pole in the right-half plane. Since the
.

will

function 1/Z(s)

is

positive real if Z(s) is positive real,

it is

obvious that there

cannot be any zeros in the right-half plane also.

Theorem D.6.

Only simple poles with

positive real residues

can

exist

on the

ya> axis.

As a consequence of the
may exist on they'd) axis if n =

Proof.

poles

that the pole

and real.

is

Proof.
real

when

If

1,

and ^

simple and the condition

It is readily

seen that zeros

The poles and

Theorem D.7.

derivation of

on

seen that

it is

implies

implies that the residue is positive

they' to axis

must also be simple.

zeros of Z(s) are real or occur in conjugate pairs.

a complex pole or zero

is real.

<j>

Theorem D.5,

= 0. The condition n

exists without its conjugate, Z(s)

As a result of this theorem and Theorem D.S,

cannot be

it is

seen that

both the numerator and denominator polynomials of Z(s) must be Hurwitz.

The highest powers of the numerator polynomial and the

denominator polynomial of Z(s) may differ by at most unity.
Proof. Let Z(s) be written as
Theorem D.8.

+ fln-iJ"-1 +
m + bn.lS"
*>m*
?"

Z{s)

+ OyS + a.
+
+b

Pis)

Q(s)

Network

492

anal/sis

and synthesis

FIG. D.2

m n ^ 2, when s = , Z() will have a zero of order 2 or greater at s = oo,

which is on the/w axis. Similarly, if n - m 2, then, at s = , Z(s) will have a
pole of order 2 or more at * = oo. Since Z(s) cannot have multiple poles or
zeros on the/a> axis, these situations cannot exist; therefore \n m\ <.\.
If

Theorem D.9. The lowest powers of P(s) and Q(s) may differ by at most unity,
Proof. The proof is obtained as in Theorem D.8 by simply substituting 1/s
for s and proceeding as described.
Theorem D.10.

A rational function F(s) with real coefficients is positive real if:

(a) F(s) is analytic in the right-half plane.

(b) If F(s)

has poles on

they'a> axis,

they must be simple and have

real, positive

residues.
(c)

Re F(ja>) ^

for all

o>.

We need only show that these three conditions fulfill the same requirements as Re Z(s) ^ for Re s ^ 0. We will make use of the minimum modulus
Proof.

states that if a function is analytic within a given region, the

min imum value of the real part of the function lies on the boundary of that
region. The region with which we are concerned is the right-half plane which is
bounded by a semicircle of infinite radius and the imaginary axis with small
indentations for theyVu axis poles. If the minimum value on theya) axis is greater
than zero, then Re Z(s) must be positive over the entire right-half plane (Fig. D.2).

theorem which

appendix

An

aid to

of

E.I

the improvement

filter

approximation

INTRODUCTION

The introduction of an additional pole and zero in the second quadrant

of the complex frequency plane, and at their conjugate locations, can give
amplitude or phase corrections to a filter approximant over some desired
band of frequencies without significantly changing the approximant at
other frequencies. However, a cut-and-try procedure for finding the best
positions for such a pole-zero pair can be tedious.
visual aid is presented
herein which reduces the amount of labor required to make modest
corrections of this type.
Constant phase and constant logarithmic gain contours for the correction by a pole-zero pair1 are plotted on transparent overlays. One of these
may be placed over a suitably scaled sheet of graph paper representing the
complex frequency plane. Then the pair-shaped phase and gain corrections along the jm axis are indicated by the intersections of the overlay

contours with this axis. Corrections which best reduce the errors in the
original approximant are then sought by variation of the overlay position
and orientation. Either phase or amplitude may be corrected. However,
it is not always possible to simultaneously improve both the phase and the
amplitude characteristics of an approximant by a single pair-shaped
correction.

F. F.

Kuo and M.

CT-9, No.
1

4,

Karnaugh, reprinted from the

December

IRE Transactions on Circuit Theory,

1962, pp. 400-404.

This will be called, hereafter, a pair-shaped correction.

493

Network

494

E.2

analysis

and synthesis

CONSTANT LOGARITHMIC GAIN CONTOURS

Suppose we begin with a transfer function G(s) with certain deficiencies

its amplitude or phase response. Let us consider the transfer function
of a corrective network G^s) such that the product

in

GAs)-GM<K)
will

CE.1)

have better gain or phase characteristics. For the purposes of

we will restrict G^s) to have the form

this

paper,

O^-C <'-)<'->
p)(s

(s

(E.2)

p)

where C is a constant, and q and p are a zero and a pole, respectively, in

the second quadrant of the complex frequency plane. If the correction
p
is to be applied at sufficiently high frequencies such that s qo*s
then

GMcsZtzA

(E.3)

s-p

Let us consider the effect when the pole-zero pair in Eq. E.3 is used to
augment any given rational function in the complex frequency plane.

The added

gain, in decibels,

D=
where we have neglected the

then

For

due to

20 log10

effect

fc

-q
s- p
s

of the constant
(lO)*'

20

E.3. If

we let
(E.5)

circle

with inverse points* at q and

is

kjp^q\
II

its

C in Eq.

(E.6)

P
and

(E.4)

s-p

fixed k, this is the equation of

p. Its radius

this pole-zero pair is

center

is

- *"l

(E7)

at

Press, Oxford,
*
E. C. Titchmarsh, The Theory of Functions, Oxford University
England, second edition 1939, pp. 191-192.

Appendix E

495

Let
(E.9)

Then

5
II

(E.10)

k*\

(E.ll)

Here, s
is e. It is

is the midpoint between the pole and zero, and their separation
easy to see from Eq. E.l 1 that the center of each circle of constant

gain

is externally collinear with the pole and zero.

For the purpose of
drawing the family of constant gain contours, we may let s be the origin
of coordinates and the scale factor e may be set equal to unity.
Furthermore, only half the pattern need be drawn because the function
D has negative symmetry with respect to a reflection about the perpendicular bisector of the pole-zero pair.

E.3

CONSTANT PHASE CONTOURS

Figure E.la represents a pole at/>, a zero at q and an arbitrary point

complex frequency plane. When the pole and zero are used to
correct a given phase characteristic, the added phase at s is
s in the

4>

9,

(E.12)

where BQ and 6 9 are measured with respect to a single arbitrary reference.

In Fig. E.l A, a circle has been drawn through p, q, and s. Angle ^ is

FIG. E.l. Derivation of constant phase contours.

Network

496

and synthesis

analysis

equal to one half the subtended arc ps'q. Therefore, the arc qsp is a
constant phase contour. The angle at the center of the circle between cp
and the perpendicular bisector of chord pq is also equal to
All of the
<f>.
circular contours of constant phase have their centers on this perpendicular
bisector.

Note that minor arc ps'q

is

also a contour of constant phase, but the

phase angle

is

negative, as indicated

<f>

(E.13)

by the clockwise rotation from s'p to s'q. The

is herein taken to be counterclockwise.

convention for positive rotation

E.4

CONTOUR DRAWINGS

Sets of constant phase

drawn

in Figs. E.2

and

and constant logarithmic gain contours are

The curves are symmetric about the zero-

E.3.

decibel line, except that the gain curves are of opposite sign.

Therefore,

only one half of each figure has been drawn.

7.5*
=^5*

-K s

~~yf

/i?\

9'

v>

I
10*

^.s\

lb"

30
i

Ills

2\

1.71 1

5|

]l

|] .2

\
1

0.7

0.6

'

TO*

20

'

lb"

10"

9*

7.5"

^v
*Sf

v^jy
-J'l

rV.

FIG. E.2. Constant amplitude and phase contours.

db

0.5

Appendix E

497

FIG. E.3. Constant amplitude and phase contours.

The

zero-decibel line

is

joining the pole and zero.

the perpendicular bisector of the line segment

It is a gain contour of infinite radius.

through the pole and zero is a zero phase contour. Together,

form a useful reference system. The signs of the
phase and gain in the four quadrants formed by these axes are shown in

The

line

these perpendicular axes

Fig. E.4.

The corrections in

Figs. E.2

and E.3 do not carry algebraic signs because

Network

498

analysis

and synthesis
Zero gain
.1

G-

G+

<t>-

-x^-*-Zero phase

G-

G+

<t>+

FIG. E.4. Signs of the gain

G and phase

<f>

corrections.

only half of the symmetrical pattern has been drawn. The signs may be
obtained from Fig. E.4, and are important in selecting the orientation of
the pole-zero pair.

E.5

CORRECTION PROCEDURE

In correcting a given approximant, it is first necessary to plot the desired

magnitude and/or phase corrections versus frequency. When the contours
of Fig. E.2 or E.3 are overlaid on a second sheet representing the complex
frequency plane, the contour intersections with the ja> axis indicate the
corrections that will actually be realized.

There are several variables at the designer's disposal. The first is the
distance between the correcting pole and zero. Since the scale of the
contours in Figs. E.2 and E.3 is not specified, the frequency scale of
the underlying complex plane determines the distance between the pole
and zero. The second design variable is the location of the center of the
pole-zero pair. The distance of the center from a given band on theyeo
axis determines the

magnitude of the correction.

is the orientation of the pole-zero pair. Figure E.4
shows how the orientation affects the gain and phase corrections. As a
simple example, suppose one wishes to have zero phase correction at
co = 1.0, negative phase correction above and positive correction below
that frequency. The attack would be to point the zero phase axis at
co = 1.0 with the pole nearest to they'to axis. If it is desired to have equal
phase correction above and below co = 1.0, the zero phase axis should be
oriented parallel to the real axis of the complex frequency plane. If one
wishes to have more phase correction above co = 1.0 and less below, the

The

third variable

Appendix E

499

zero phase axis should be rotated clockwise with respect to the a (real)
axis.

We thus see that by varying the frequency scale of the complex frequency
plane, the position of the center of the pole-zero pair,

and the orientation

of the pole-zero pair, the pair-shaped correction can be made to approxi-

mate the desired correction.

It must be emphasized that the method suggested herein is an aid to
cut-and-try correction. As such, it is easier to use the method than to
precisely set down rules for applying it. However, a few rather general

may be helpful.
Unless the pole and zero both lie on the real axis, one must remember
that another pole and zero are located at conjugate positions. The contributions from both pole-zero pairs may be added algebraically. In most
practical cases, the desired correction will have a band-pass character.
statements

Therefore, only one pole-zero pair will normally contribute significantly

at

any frequency.

The shape and magnitude of the

way
The broadness

desired correction will dictate the

intersect the correction contours.

which the/co axis must

of the desired correction will dictate the proper scaling of the jco axis.
Usually, only a few trials are needed to fix the pole and zero locations for

in

the best

fit.

be found that a worthwhile correction can be made in either the

phase or the gain characteristic. Only fortuitously can they be improved
simultaneously by a single pair-shaped correction.
It will

Example E.1. The amplitude response of a third-order Butterworth filter is

given by the solid curve in Fig. E.5. It is desired to steepen the gain roll-off
near the cutoff frequency o c = 1 This is done by increasing the gain just below
o) = m e and decreasing it above that frequency. Figure E.6 illustrates the type
of correction desired. Figure E.7 shows a pole-zero pair that achieves this type
of correction. The gain at to = 1 remains unchanged by this particular choice.
Other pole-zero pairs that "aim" the zero-decibel line at to = 1 but give asymmetrical corrections about that point might also be used. The dashed curve in
Fig. E.S shows the corrected gain.
.

different pole-zero pair, also shown in Fig. E.7, has been chosen to minimize
the deviation of the slope of the phase response from its slope at to = 0. This
pair-shaped correction decreases the phase for m < 0.7 and increases it for
co

> 0.7.

Figure E.8 shows the deviation of the phase responses from linear phase.
It is clear from Figs. E.5 and E.8 that the gain corrected approximant has a
poorer phase response than the original, while the phase corrected approximant

has a gentler gain roll-off than the original.

500

Network

anal/sis

and synthesis

0.2

0.5

0.3

Frequency,

0.7

1.0

1.5

2.0

FIG. E.5. Amplitude response of corrected and uncorrected Butterworth

This will not surprise the experienced

filter

designer.

It is possible,

filters.

however,

both gain and phase by using two pairshaped corrections. A useful approach to this objective lies in localizing the
gain correction further out of band, and the phase correction further in-band
than in the separate corrections just discussed. This can be done by shifting
the pole-zero centers to higher or lower frequencies and also by experimenting
with nonsymmetrical corrections.
to achieve moderate corrections in

FIG. E.6. Amplitude correction to steepen

fall-off

of third-degree Butterworth

filter.

Appendix E

Loss

1.2
1

correction

o
1.0

Butterworth

poles

0.8

X o
Phase

0.6

correction

0.4

0.2

_L

J_

-1.0 -0.8 -0.6 -0.4 -0.2

\

-0.2

\
\

-0.4

-0.6

x o

-0.8

-1.0

X
-1.2

FIG. E.7. Poles and zeros of the original

filter

and correction equalizers.

20
>

10
With >hase :orrect on

s -10

^'

Tl ird-orc er

^^v

/
//

//
/

/
/

zL

Biitterwo rvn'/

\\ 1
\

-20

>

/w th amp litude

correct on

-30

-40
0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

Frequency, rad/sec

FIG. E.8. Phase deviation from slope at zero frequency.

501

Network

502

E.6

anal/sis

and synthesis

CORRECTION NETWORK DESIGN

few words are in order concerning the synthesis of the equalizer from
the pole-zero pair obtained by the method described.
In order to provide optimum power transfer and to facilitate cascading
the correction network with the original
should be of the constant-resistance type.

to the bridged-T network in Fig. E.9,

given as

the correction network

filter,

We will restrict

whose voltage

our discussion

transfer function

G(s)

provided the network

equation

is

is

(E.14)

of the constant resistance type as given by the

Z (j)Z2(i) =
1

For normalization purposes, we

will let

(E.15)

R =

Q.

Let the pole-zero correction be written in general as

G(s)
Since the correction has

+as+a
fc(s* + b lS + b
s

minimum

phase,

we know

addition, since the d-c gain cannot exceed unity,

We

(E.16)
)

kb

that {a ( , b t

^a

^ 0}.

In

can express the impedance Z^s) in terms of G(s) in Eq. E.14 as

Zl(s)

_L_ 1= (

+ (kb, - ajs + (fcb ~,

T~,
s + a ts + a

l)s*

G(s)

Since Z^s) must be positive real, the coefficients must

so that

all

a)

(Ai')

be nonnegative

k- ^0
1

kbi
kb

<*i

^
^

(E.18)

Zi

So

Vi

So

So

Z2

So

FIG. E.9. Bridged-T network.

v2

Appendix E

503

-ju
FIG. E.I0. Poles and zeros of Zt(s).

Moreover, in order for a biquadratic driving-point immittance to be

positive real, the following condition

Details concerning the synthesis of

Let us plot the pole-zero pair of

must apply. 8

Z^j) are

also given in Seshu's paper.

Zt{s) as in Fig. E.10. We can represent

< and 0,
Rack* has shown that in

the locations of the poles and zeros in terms of the polar angles

and

their distances

from the

order for in-band loss to be

origin,

n p and n t

finite

^<2

(E.20)

In addition he has shown that for an unbalanced bridged-T circuit, if

a = max [0M , <f>9] then the larger the angle a, the larger the in-band loss.
In particular, should be less than 70 to restrict the in-band loss to
reasonable proportions.

If

one considers other network configurations

*S. Seshu, Minimal Realizations of the Biquadratic

December 1959, 345-350.

Trans, on Circuit Theory, CT-6,

4

A.

J.

Rack, private communication.

Minimum

Function,

IRE

Network

504

analysis and synthesis

such as lattice networks, it is conceivable that Rack's restrictions might

5
be relaxed. For lattice network design, Weinberg's method is applicable,
although here again, the in-band loss restrictions are severe.

E.7

CONCLUSION

We

have presented here a simple visual aid to the correction of the

amplitude or phase response of filters. The method has the advantage of
facilitating the commonly used cut-and-try approach to this problem. It
has the disadvantage, in many cases, of failing to provide simultaneous
amplitude and phase corrections in a single step.
1

L. Weinberg,

"RLC lattice

networks," Proc. IRE, 41, September 1953, 1139-1144.

Bibliography

SIGNAL ANALYSIS
Goldman,

S.,

McGraw-Hill,
Frequency Analysis, Modulation and Noise,
'

New
Javid,

York, 1948.

M. and

E. Brenner,

McGraw-Hill,

^w/y^w, Transmission and Filtering of Signals,

New York,

1963.

Lathi, B. P., Signals, Systems

New

and Communications, John Wiley and

5>ons,

York, 1965.

Lighthill,

Functions,

M. J., Introduction to Fourier Analysis and Generalized

Press, New

York, 1958.

Cambridge University
Circuits, Signals and
Mason, S. J. and H. J. Zimmerman, Electronic
1960.
Systems, John Wiley and Sons, New York,
McGraw-Hill, New
Applications,
Its
and
Integral
Papoulis, A., The Fourier
York, 1962.

Rowe, H.

E., Signals

and Noise

in

Communications Systems, D. van

Nostrand, Princeton, N.J. 1965.

Modulation and Noise,
Schwartz, M., Information Transmission,
Hill, New York, 1959.

McGraw-

NETWORK ANALYSIS
Allyn and Bacon, Boston,
Balabanian, N., Fundamentals of Circuit Theory,
Systems, Addison-Wesley,
Bohn, E. V., The Transform Analysis of Linear

Reading, Massachusetts, 1963.

Brenner, E. and

M. Javid, Analysis of Electric Circuits, McGraw-Hill,

York, 1959.
505

New

Network

506

analysis and synthesis

Brown, R. G. and J. W. Nilsson, Introduction to Linear Systems Analysis,

John Wiley and Sons, New York, 1962.
Brown, W. M., Analysis of Linear Time-Invariant Systems, McGraw-Hill,
New York, 1963.
Carlin, H. J. and A. Giordano, Network Theory, Prentice-Hall, Englewood
Cliffs, N.J., 1964.

W.

Cassell,

L.,

Linear Electric Circuits, John Wiley and Sons,

New York,

1964.

Chen,

W.

H., The Analysis of Linear Systems, McGraw-Hill,

New

York,

1963.

dePian, L., Linear Active Network Theory, Prentice-Hall, Englewood

Cliffs, N.J. 1962.
Friedland, B., O. Wing,

and R. B. Ash, Principles of Linear Networks,

McGraw-Hill, New York, 1961.
Gardner, M. and J. L. Barnes, Transients in Linear Systems, Vol. 1, John
Wiley and Sons, New York, 1942.
Guillemin, E. A., Introductory Circuit Theory, John Wiley and Sons,

New

York, 1953.
Guillemin, E. A., Theory ofLinear Physical Systems, John Wiley and Sons,

New York,

1963.

Harman, W. W. and D. W.
McGraw-Hill,

New York,

Lytle, Electrical

and Mechanical Networks,

1962.

W. H., Jr. and J. E. Kemmerly, Engineering Circuit Analysis,

McGraw-Hill, New York, 1962.
Huelsman, L. P., Circuits, Matrices and Linear Vector Spaces, McGrawHill, New York, 1963.
Kim, W, H. and R. T. Chien, Topological Analysis and Synthesis of Communication Networks, Columbia University Press, New York, 1962.
Ku, Y. H., Transient Circuit Analysis, D. Van Nostrand, Princeton, N.J.,
Hayt,

1961.

Legros, R. and A. V.

New

W.

R. and
York, 1952.

LePage,

Martin, Transform Calculusfor Electrical Engineers,

J.

Prentice-Hall, Inc.,

Englewood

S. Seely,

Cliffs, N.J., 1961.

General Network Analysis, McGraw-Hill,

Ley, B. J., S.G. Lutz, and C. F. Rehberg, Linear Circuit Analysis,

Hill,

New York,

Lynch,

W. A. and

Hill,

New York,

McGraw-

1959.
J.

G. Truxal, Introductory System Analysis, McGraw-

1961.

Paskusz, G. F. and B. Bussell, Linear Circuit Analysis, Prenctice-Hall, Inc.

Englewood
Pearson, S.

I.

and Sons,

Cliffs, N.J., 1964.

and G.

J.

New York,

Maler, Introductory Circuit Analysis, John Wiley

1965.

Bibliography
Pfeiffer, P. E.,

Reza, F.

507

New York, 1961.

McGraw-Hill, New

Linear System Analysis, McGraw-Hill,

M. and

S. Seely,

Modem Network Analysis,

York, 1959.
Sanford, R.

S.,

Physical Networks, Prentice-Hall, Englewood

Cliffs, N.J.,

1965.

Schwarz, R. J. and B. Friedland, Linear Systems, McGraw-Hill, New York,

1965.
Scott,

R.

E., Elements

of Linear

Circuits,

Addison-Wesley, Reading,

Massachusetts, 1965.
Seely, S.,

Seshu, S.

Sons,

Dynamic Systems Analysis, Reinhold, New York, 1964.

and N. Balabanian, Linear Network Analysis, John Wiley and

New York,

1959.

H. H., Electrical Engineering Circuits, Second Edition, John

Wiley and Sons, New York, 1965.
Van Valkenburg, M. E., Network Analysis, Second Edition, Prentice-Hall,
Englewood Cliffs, N.J., 1964.
Weber, E., Linear Transient Analysis, John Wiley and Sons, New York,

Skilling,

1954.

Zadeh, L. A. and C. A. Desoer, Linear Systems Theory, McGraw-Hill, New

York, 1963.

NETWORK SYNTHESIS
Balabanian, N., Network Synthesis, Prentice-Hall, Englewood

Cliffs, N.J.,

1958.

Bode, H. W., Network Analysis and Feedback Amplifier Design, D. Van

Nostrand, Princeton, N.J., 1945.
Calahan, D. A., Modern Network Synthesis, Hayden, New York, 1964.
Chen, W. H., Linear Network Design and Synthesis, McGraw-Hill, New

York, 1964.
Geffe, P. R., Simplified

Modern

Filter Design,

John F. Rider,

New York,

1963.

Guillemin, E. A., Synthesis of Passive Networks, John Wiley and Sons,

New York,

1957.

Guillemin, E. A., The Mathematics of Circuit Analysis, John Wiley and

Sons, New York, 1949.

Hazony, D., Elements of Network Synthesis, Reinhold, New York, 1963.
Kuh, E. S. and D. O. Pederson, Principles of Circuit Synthesis, McGrawHill, New York, 1959.
Matthaei, G. L., L. Young, and E. M. T. Jones, Microwave Filters,
Impedance-Matching Networks and Coupling Structures, McGraw-Hill,
New York, 1964.

Network

508

Saal, R.,

analysis

and synthesis

Der Entwurf von

Tiefpdsse, Telefunken

Skwirzynski,

J.

Filtern mit Hilfe des Kataloges Normierter

GMBH,

1961.

K., Design Theory

and Data for

Electrical Filters,

D. Van

Nostrand, Princeton, 1965.

Storer, J. E., Passive

Truxal,
Tuttle,

J.

D.

New York, 1957.

New York, 1955.
1, John Wiley and Sons, New York,

Network Synthesis, McGraw-Hill,

G., Control System Synthesis, McGraw-Hill,

F.,

Network Synthesis, Vol.

1958.

Van Valkenburg, M.

E., Introduction to Modern Network Syntheses,

John Wiley and Sons, New York, 1960.
Weinberg, L., Network Analysis and Synthesis, McGraw-Hill, New York,

1962.

Yengst,

W. C, Procedures of Modern Network Synthesis, Macmillan, New

York, 1964.

Name
Eisenman, R. L., 469
Elmore, W. C, 390
Ende, F., 381

Aaron, M. R., 448

Aitken, A. C, 469
Angelo, E. J., 268
Ash, R. B., S06

Faddeeva, V. N., 469

Faddev, D. K., 469
Fano, R. M., 448
Feshbach, H., 86
Foster, R. M., 321
Franklin, P., 198
Friedland, B. O., 506
Fujisawa, T., 458
Fukada, M., 381

Balabanian, N., 505, 507

J. L., 506
Bashkow, T. R., 283, 438
Bellman, R., 469
Bode, H. W., 221, 507
Bohn, E. V., 505
Brenner, E., 505
Brown, R. G., 506
Brown, W. M., 506
Bubnicki, Z., 283
Budak, A., 395
Bussell, B., 506

Barnes,

Gantmacher, F. R., 469

W. H., Jr., 506
Geffe, P. R., 507
Giordano, A., 506
Goldman, S., 137, 505
Hayt,

Calahan, D. A., 507

Goldstone, L. O., 153

Guillemin, E. A., 296, 299, 506, 507.

428
Cassell, W. L., 506
Cauer, W., 324
Chen, W. H., 506
Chien, R. T., 506
Carlin,

H.

I.,

Darlington,

S.,

Harman, W. W., 506

W. H., Jr., 506

Hazony, D., 507
Hobson, E. W., 471
Hohn, F. E., 469
Huelsman, L. P., 261, 469, 506
Hayt,

431, 457

Davis, H. F., 51

de Pian,

L.,

506
Jahnke, E., 381
James, R. T., 389
Javid, M., 505

Desoer, C. A., 507

Dirac, P. A. M., 33, 470, 471

Dutta Roy,

S.

Index

C, 283
509

Name

510

Jones, E.

Jordan, E.

Index

M. T., 507
C, 413

Reza, F. M., 507

Rowe, H.

E.,

505

27

Justice, G.,

Saal, R., 441

Karnaugh, M., 493

W.

Kautz,

Sands, M., 390

H., 47, 367

Sanford, R.

Kemmerly, J. E., 506

Kim, W. H., 506
Ku, Y. H., 506
Kuh, E. S., 507
Kuo, F. F., 279, 393, 493
Larky, A.

Semmelman, C.

Lathi, B. P., 505

Le

Seshu,

Corbeiller, P.,

Lighthill,

R.,

M.

.,

506

65, 136, 471, 473,

Lutz, S. G., 506

Lynch,
Lytle,

W.

447

Skwirzynski,

506

J.,

L.,

503, 507

Skilling,

Leichner, G. H., 279

Ley, B.

S.,

H. H., 100, 104, 106, 181, 507

J. K., 508
So, H. C, 449, 450
Storer, J. E., 508
Szentirmai, G., 441, 453

469

Legros, R., 506

Le Page, W.

507

Schwartz, M., 505

Schwarz, R. J., 507
Scott, R. E., 507
Seely, S., 506, 507

262

I.,

S.,

Saraga, W., 443

Schwartz, L., 470, 471

A., 506

D. W., 506

Maler, G. J., 506

Marcus, M., 469
Martin, A. V. J., 506
Mason, S. J., 505
Matthaei, G. L., 507
Mine, H., 469
Morse, P. M., 86
Nering, E. D., 469
Nilsson, J. W., 506

505
Tellegen, B. D. H., 272

Temple, G., 34, 471

Terman, F. E., 236
Thomson, W. E., 384
Titchmarsh, E. C, 494
Tompkins, C. B., 447
Tropper, A. M., 469
Truxal, J. G., 506, 508
Tuttle, D. F., 508
Ulbrich, E., 441
Jr., 293
Valkenburg, M. E., 135, 263, 296,

Valley, G. E.,

Van

297, 325, 329, 331, 344, 347, 376,

O'Meara, T. R., 283

Orchard, H. J., 385

A. C, 291

507, 508

Von

Weiss, H., 469

Papoulis, A., 379, 381, 505

Walker, F., 283

Wallman, H., 293

Paskusz, G. F., 506

Weber,

Paley, R. E.

506
Pederson, D. O., 507
Perlis, S., 469
Pfeiffer, P. E., 507
Pipes, L. A., 469

Pearson, S.

Rack, A.

J.,

I.

E., 507
Weinberg, L., 400, 435, 504
Widder, D. V., 136
Wiener, N., 291

Yengst,

W. C, 508

Young,

L.,

507

503

Raisbeck, G., 8, 294

Rehberg, C. F., 506

Zadeh, L. A., 507

Zimmerman, H., 505

Subject Index

ABCD

parameters, 262

Admittance, driving-point,
All-pass network, 221, 357

15,

187

Cauchy-Riemann equations, 485

Cauer ladder expansion, 324
Causality, 9, 290
Characteristic equation, 77

Amplifier, 11

impulse response of, 201

Amplitude response, 212, 342

Characteristic impedance,

computer program for, 450

evaluation by vector method, 215
Amplitude spectrum, 3
Analytic function, 485
Approximation problem, 17, 365
Available gain, 418
Available power, 418

Cbebyshev

414

Characteristic value, 78
filter,

373

approximation, 366
locus of poles, 376, 378
tables of, 373, 400, 435
transient response, 392
Chebyshev polynomial, 373
Circuit, bridged-T 258, 275,

502

double-tuned, 238

Band-elimination

filter

255
229
symmetrical, 259

reciprocal, 8,

transformation,

single-tuned,

409
Band-pass filter transformation, 407
Bandwidth, half-power, 235
spectral, 67, 389
Bessel

filter,

383, 435

phase response of, 387

tables of,

Coefficient of coupling, 122

Compensation theorem, 181

Complementary function, 82
variables, 481

Complex

analysis,

400, 435

483
484
487

differentiation,

392
Bessel polynomial, 386, 395
transient response of,

integration,

programs,

438,

439,

Biquadratic immittance, 305, 503

Computer

Black box, 15

441, 447 448, 450 453, 457

Constant-resistance network, 352, 502

Bode plots, 221

Bounded real function, 419
Break frequency, 225
Bridge circuit, 285, 354
Butterworth

filter,

368, 434, 500

amplitude response of, 369, 387, 394,

500
pole locus of, 371
step response of, 391, 394
tables of, 372, 400,

Contour

18,

487

integration,

Controlled source, 268

Convergence in the mean, 49, 51
Convolution integral, 197
Crest factoT, 27
Critical coupling, 242
Current source, 12
Cutoff frequency, 17, 225

435

Damping

factor,

225

d-c value, 25

Canonical form, 325, 400, 433

Capacitor, 13, 103, 176
Cauchy integral theorem, 489
Cauchy residue theorem, 487

Delay, 32, 212, 245, 384

distortion,

time, 388

511

245

Subject Index

512

Fourier transform, phase spectrum, 64,

Delta function, see Unit impulse

Delta-wye transformation, 257
Differential equation, 75, 145

65
properties of, 67

forcing function of, 75

homogeneous, 75, 76
integrodifferential equation, 91,

145

76
nonhomogeneous, 75
ordinary, 76

linear,

simultaneous, 93, 146

Differentiator,

193

18,

11,

Digital computer, 18,

438

Distributions, theory of,

Dot reference

470

for transformer,

123

Doublet, 40, 479

Duhamel superposition integral, 201

Duty

cycle,

symmetry conditions, 68
Free response, 106
Frequency, angular, 2
complex, 4
domain, 14, 134, 367
forced, 192
natural, 192
Frequency normalization, 18, 402
Frequency transformation, 18, 404

26

Energy density, 71
Energy spectrum, 71
Equal ripple approximation, see Chebyshev approximation
Essential singularity, 486

Gain contours, logarithmic, 494, 496

Gaussian filter, 291
Generalized functions, 34, 64, 470
g parameters, 286
Green's function, 86
Gyrator, 272, 287
Hall effect, 273
High-pass filter transformation, 405
Hurwitz polynomial, 294, 316 347

Hybrid (h) parameters, 260

Hybrid matrix, 261, 449

Excitation, 1

Faraday's law of induction, 122

Filter approximation, 368, 373,

379,

Ideal low-pass

filter,

457
441, 457

Filter design, 17, 365, 397, 433,

computer programs

for,

driving-point,

Flux linkage, 122

Forced response, 106

transfer,

1,

46,

15, 253,

50

amplitude spectrum, 57
complex form, 55
cosine series, 53
evaluation of coefficients, 52, 58

188

Impulse function, see Unit impulse

Impulse response, 44, 85, 111, 194
Incident power, 416
Incidental dissipation,

13, 77,

Initial conditions,

phase spectrum, 57

Insertion loss, 429

Fourier transform, 1, 63
amplitude spectrum, 64, 65
inverse,

63

64

of unit impulse, 65
of unity, 68

86 107, 176

value theorem, 165

Initial

conditions, 53

276

Inductor, 13, 104, 176

orthogonality conditions, 49

discrete, 56,

315

matrix, 254

transformer, 263

Foster network, 321, 325

Fourier series,

424

Immittance, 15, 315

Impedance, 15, 176

Final conditions, 106, 109

Final value theorem, 165

symmetry

292, 367

Ideal transformer, 263, 273, 275,

383, 493

filter synthesis,

Insertion

power

431

ratio,

430

voltage ratio, 429

Integrator, 11, 18, 193
Integrodifferential equation, 91, 145

Kirchhoffs laws, 100, 104

513

Subject Index

Ladder network, 279, 346, 347, 453

Laguerre polynomial, 48
Laplace transform, 1, 134
definition of, 135
inverse,

136

Node equations,
Norm, 47

255

106,

Normalization, frequency, 402

magnitude, 402
Norton's theorem, 180, 185

properties of, 137

Optimization techniques, 447

Optimum (L)

486
L-C immittance, 315
properties of, 315
synthesis of, 319
series,

379

filter,

amplitude response of, 380

polynomials, 382

Least squares, principle of, 49, 366

Legendre polynomial, 381

Linear phase filter, see Bessel

Linear system, 8
derivative property,

parameters,

254, 343

uses of, 144

Lattice network, 285, 354, 503

Laurent

impedance

Open-circuit

table of, 168

filter

Orthogonal set, 47
Orthonormal set, 47
Overcoupling, 242
Overshoot, 388, 392

Paley-Wiener criterion, 291

Parseval's equality,

ideal elements, 12

50

theorem, 71

ideal models, 10

Partial fraction expansion, 148

MAC,

Project,

conjugate poles, 150

multiple poles, 151

448

Machine-aided design, 448

Magnitude normalization, 18, 402

Matrix algebra, 461
definitions, 462
operations, 464
references, 469
Maximally flat response, see Butterworth filter
Mean squared error, 48

Minimum

inductance

transformation,

443

Minimum modulus theorem, 492

Modulation, amplitude, 70
Monotonic

filter,

see

Optimum

filter

Multiple-access computer, 448

Mutual inductance, 122

real poles, 149

Particular integral, 82

Passive network, 8

Peaking

233, 241

circle,

Peak-to-valley ratio, 251

Phase contours, logarithmic, 495

Phase response, 212, 343

computer program for, 450
evaluation by vector method, 215
linearity of, 213, 383
Phase

shift, 1

distortion,

213

minimum, 220, 346

spectrum,

3,

57

Phasor, 5
Plancheral's theorem, 71

Negative impedance converter (NIC),

261
Network, analysis, 7, 100, 175, 253
linear, 8,

n-port,

449
100

reciprocal, 8, 100,

255

symmetrical, 259
synthesis, 290, 315, 341, 397, 431

time-invariant, 9,

Pole-zero diagram, 156

Port, 12, 253
Positive real

100

passive, 8,

Pole, definition of, 155

100

(p.r.)

function,

16, 299,

490
Power, average, 301, 315
available, 418
Propagation constant, 414
Proportionality, principle of, 8

Pulse transmission, 389, 392

514
Q,

Subject Index
Spectra, continuous, 3, 63

236

circuit, 233,

57

line (discrete), 3,

Ramp

Stability,

function, 31

R-C admittance,
R-C impedance,

properties of, 331

properties of, 325

290

marginal, 293
Steady-state solution, 106, 109

Steepest descent method, 447

synthesis of, 329

Realizability conditions,

342,

116,

16, 290,

301,

Step function, see Unit step

Step response, 44, 45, 85, 111, 194

490

for driving-point functions, 301

filter

for transfer functions, 342

Reciprocal network,

8,

255

Rect function, 65
Reference impedance factor, 415
Reflected parameter, 415
Reflected power, 416
Reflection coefficient, 416, 421, 458
Remainder function, 308
Residue, definition of, 162
evaluation by vector method, 162
Residue condition for two-ports, 344
Resistor, 13, 103, 175
Response, critically damped, 118
overdamped, 118
underdamped, 119
Ringing, 388
Rise time, 388
R-L admittance, properties of, 325
synthesis of, 329
R-L impedance, properties of, 331
synthesis of, 331
Sampled-data system, 142
Sampler, 18, 142
element, 416

R-L-C

functions, 333

transfer functions, 347, 352

System function,

14, 187,

admittance
circuit,

Signal, continuous,

decomposition
parts, 21

parameters,

248, 400

23

into

even

20
46

symmetrical, 21, 54

Signum (sgn) function, 30

Thevenin's theorem, 180

Time constant, 24
Time delay, 32, 212, 245, 384, 388
Time delayer, 11
Time domain, 14, 134, 366
Time-invariant system, 8

Transfer function,

14,

486

16,

188,

266,

341, 352

admittance, 188, 347

impedance,

Transformed

16,

188, 347

175
Transformer. 122. 177
ideal, 263
circuit,

filters,

388

and odd

common

emitter,

Transmission

coefficient, 421,

Transmission

line,

286
457

413, 425

Transmission matrix, 262

Transmittance, 16
Two-port, 12, 16, 254, 264
cascade connection, 271
equivalent circuits of, 268, 271
matrix representation, 264
parallel connection of,
series

273

connection of, 275

Sine function, 65
Singularity,

194

Terminals, 12

Transistor,

257, 344

periodic, 20,

functions, 325, 329, 331

functions, 325, 329, 331

Transient solution, 106

Settling time, 388

deterministic,

397

synthesis,

functions, 315, 319

unity coupled, 125

matrix, 420

Shunt peaked

L-C
R-C
R-L

Transient response of low-pass

Scattering parameters, 415, 419

Short-circuit

Synthesis procedures elementary, 308

Undercoupling, 241

515

Subject Index

Undetermined coefficients, method

g2
Uniform loading, 278
Unit coupled transformer, 121

of,

Unit impulse (delta) functions, 33, 43,

47(j
derivative of,

Voltage ratio, 16, 188

Voltage source, 12
Voltage standing wave

ratio,

428

y-parameters, 257, 344

see also Short-circuit parameters

40

properties of, 34, 36, 476,

function, 31

Unit ramp
Unit step function, 28, 474
derivative of, 34, 37, 475

480

z -parameters, 254 343

see also Open-circuit parameters

z-transform, 142

Zero, definition of , 155

Zero of transmission, 345

o
This Wiley international Edition

gram of paperbound

text

is

part of a continuing pro-

books especially designed

dents and professional people overseas.

reprinting of the original

hardbound

It

is

m
o
O
z
D
m
g
o

for stu-

an unabridged

edition,

which

is

also

available from your bookseller.

Wiley International Editions

include

titles in

the fields of:

Chemical Engineering
Civil

Engineering

Electrical Engineering

Psychology
Agricultural Engineering

Biological Science

Metallurgy

Chemistry

Mechanical Engineering

Earth Sciences

Business

Management

(A

and Agriculture

Mechanics

Statistics

3
o

in

>
2
O

Mathematics
Physics

c/>

C/>

JOHN WILEY & SONS, INC., U.S.A.

TOPPAN COMPANY LTD., JAPAN

WILEY

TOPPAN
[P 10)