DIRE DAWA UNIVERSITY INSTITUTE OF

TECHNOLOGY
SCHOOL OF ELECTRICAL AND COMPUTER ENGINEERING

COURSE NAME: SIGNALS AND SYSTEM

ANALYSIS
COURSE CODE: ECEG2141
(Lecture Notes)
 Syllabus:
Sl.No Topics or Subtopics or Chapters

UNIT 1 1.Introduction
Characterization, Classification of signals and systems Representation/modelling of
signals and systems

UNIT 2 2. Signal Approximation

Orthogonal/orthonormal and basis functions; error to signal energies ratio.
UNIT 3 3. Fourier Series Representation Signals
Trigonometric & exponential Fourier series, frequency and power spectrums, error
to signal energies ratio.
UNIT 4 4. Singularity Functions
The impulse, step and ramp functions; discontinuous functions.
UNIT 5 5. Convolution of Continuous Time Signals
Signals as a continuum of impulses; impulse responses; analytical, graphical and
numerical evaluation of convolution
UNIT 6 6. Modeling of Electrical Systems and Ordinary Linear Differential Equations
(OLDE)
Single and multiple variable OLD equations
UNIT 7 7. Fourier Transforms and Inverse Transforms
Properties, energy spectrum, transform theorems, system functions, the Sampling
Theorem
Ideal and real sampling, Nyquist rate.
UNIT 8 8. Laplace Transforms and Inverse Laplace Transforms
(a) Properties, relationship between Fourier and Laplace transforms; system
functions;
(b) Partial fraction method, modified power series methods, evaluations from pole-
zero plots; systems response, solutions to integral-differential equations;
UNIT 9 9.Z-Transforms and Inverse Transforms
(a) Discrete signals and systems, the Z-transform & inverse transform; system
functions; difference equations and their solutions
(b) Introduction to Discrete Time Fourier Transform (DTFT)
Course Description:
This course will introduce students to mathematical descriptions of signals & systems, and
mathematical tools for analyzing and designing systems that can operate on signals to achieve a desired
effect. Classification of signals and systems; signal representation and approximation - orthogonal and
basic functions, Fourier series representation; singularity functions; convolution - analytical, graphical and
numerical computations; single and a system of ordinary linear differential equations as applied to
network analysis; the Fourier transform; the sampling theorem; forward and inverse Laplace
transformations - system function and other applications; forward and inverse Z-transformations and
applications, solving difference equations; introduction to discrete time Fourier transform (DTFT)

Textbook
[1] Analysis of Signals and Systems, Tesfaye Bayou.
References:
[1] Signals and Linear Systems, R.A. Gabel & Richard A. Roberts

[2] Signals and Linear Systems, Lathi

[3] Advance Engineering Mathematics, E.Kreyzig

UNIT I
INTRODUCTION TO SIGNALS :

1.1 INTRODUCTION

Anything that carries information can be called a signal. Signals constitute an important part of
our daily life. A Signal is defined as a single- valued function of one or more independent
variables which contain some information. A signal may also be defined as any physical quantity
that varies with time, space or any other independent variable. A signal may be represented in
time domain or frequency domain. Human speech is a familiar example of a signal. Electric
current and voltage are also examples of signals. A signal can be a function of one or more
independent variables. A signal can be a function of time, temperature, position, pressure,
distance etc. If a signal depends on only one independent variable, it is called a one-dimensional
signal, and if a signal depends on two independent variable, it is called a two-dimensional signal.

1.2 REPRESENTATION OF DISCRETE TIME SAIGNALS

Discrete-time signals are signals which are defined only at discrete instants of time. For those
signals, the amplitude between the two time instants is just not defined. For discrete time signal
the independent variable is time n, and it is represented by x(n).
There are following four ways of representing discrete-time signals:
1. Graphical representation
2. Functional representation
3. Tabular representation
4. Sequence representation
UNIT II SIGNAL APPROXIMATION USING ORTHOGONAL FUNCTIONS

Analogy Between Vectors and Signals

There is a perfect analogy between vectors and signals.

Vector
A vector contains magnitude and direction. The name of the vector is denoted by bold face type
and their magnitude is denoted by light face type.

Example: V is a vector with magnitude V. Consider two vectors V1 and V2 as shown in the
following diagram. Let the component of V1 along with V2 is given by C12V2. The component of a
vector V1 along with the vector V2 can obtained by taking a perpendicular from the end of V1 to the
vector V2 as shown in diagram:

The vector V1 can be expressed in terms of vector V2

V1= C12V2 + Ve

Where Ve is the error vector.

But this is not the only way of expressing vector V1 in terms of V2. The alternate possibilities are:

V1=C1V2+Ve1

V2=C2V2+Ve2
1/9
V2 C2V2+Ve2

The error signal is minimum for large component value. If C12=0, then two signals are said to be
orthogonal.

Dot Product of Two Vectors

V1 . V2 = V1.V2 cosθ

θ = Angle between V1 and V2

V1 . V2 =V2.V1

The components of V1 alogn V2 = V1 Cos θ =

V 1.V 2

V2

From the diagram, components of V1 alogn V2 = C 12 V2

V1 . V2

V2 = C1 2 V2

V1 . V2
⇒ C12 =
V2

Signal
The concept of orthogonality can be applied to signals. Let us consider two signals f1(t) and f2(t).
Similar to vectors, you can approximate f1(t) in terms of f2(t) as

f1(t) = C12 f2(t) + fe(t) for (t1 < t < t2)

⇒ fe(t) = f1(t) – C12 f2(t)

One possible way of minimizing the error is integrating over the interval t1 to t2.

2/9
 Signals Analysis

t2
1
∫ [fe (t)]dt
t2 − t1 t1

t2
1
∫ [f1 (t) − C12 f2 (t)]dt
t2 − t1 t1

However, this step also does not reduce the error to appreciable extent. This can be corrected by
taking the square of error function.

1 t2 2
ε = ∫ [fe (t)] dt
t 2 −t 1 t1

1 t2 2
⇒ ∫ [fe (t) − C12 f2 ] dt
t 2 −t 1 t1

Where ε is the mean square value of error signal. The value of C12 which minimizes the error, you

need to calculate
dε
= 0
dC12

d 1 t2 2
⇒ [ ∫ [f1 (t) − C12 f2 (t)] dt] = 0
dC12 t 2 −t 1 t1

1 t2 d d d
2 2 2
⇒ ∫ [ f (t) − 2f1 (t)C12 f2 (t) + f (t)C ]dt = 0
t 2 −t 1 t1 dC12 1 dC12 dC12 2 12

Derivative of the terms which do not have C12 term are zero.

t2 t2
2
⇒ ∫ −2f1 (t)f2 (t)dt + 2C12 ∫ [f (t)]dt = 0
t1 t1 2

t
2
∫ f1 (t)f2 (t)dt

If component is zero, then two signals are said to be orthogonal.

t
1
C12 = t
2 2
∫ f2 (t)dt
t
1

Put C12 = 0 to get condition for orthogonality.

t
2
∫ f1 (t)f2 (t)dt

0=
t
1

t 2
2
∫ f2 (t)dt
t
1

3/9
 t2

∫ f1 (t)f2 (t)dt = 0
t1

Orthogonal Vector Space

A complete set of orthogonal vectors is referred to as orthogonal vector space. Consider a three
dimensional vector space as shown below:

Consider a vector A at a point (X1, Y1, Z1). Consider three unit vectors (VX, VY, VZ) in the direction of
X, Y, Z axis respectively. Since these unit vectors are mutually orthogonal, it satisfies that

VX . VX = VY . VY = VZ . VZ = 1

VX . VY = VY . VZ = VZ . VX = 0

You can write above conditions as

1 a = b
Va . Vb = {
0 a ≠ b

The vector A can be represented in terms of its components and unit vectors as

4/9
 A = X1 VX + Y1 VY + Z1 VZ . . . . . . . . . . . . . . . . (1)

Any vectors in this three dimensional space can be represented in terms of these three unit vectors
only.

If you consider n dimensional space, then any vector A in that space can be represented as

A = X1 VX + Y1 VY + Z1 VZ +. . . +N1 VN . . . . . (2)

As the magnitude of unit vectors is unity for any vector A

The component of A along x axis = A.VX

The component of A along Y axis = A.VY

The component of A along Z axis = A.VZ

Similarly, for n dimensional space, the component of A along some G axis

= A. V G. . . . . . . . . . . . . . . (3)

Substitute equation 2 in equation 3.

⇒ C G = (X1 VX + Y1 VY + Z1 VZ +. . . +G1 VG . . . +N1 VN )VG

= X1 VX VG + Y1 VY VG + Z1 VZ VG +. . . +G1 VG VG . . . +N1 VN VG

= G1 since VG VG = 1

I f VG VG ≠ 1 i.e.VG VG = k

AVG = G1 VG VG = G1 K

(AVG )
G1 =
K

Orthogonal Signal Space

Let us consider a set of n mutually orthogonal functions x1(t), x2(t)... xn(t) over the interval t1 to t2.
As these functions are orthogonal to each other, any two signals xj(t), xk(t) have to satisfy the
orthogonality condition. i.e.

5/9
 t2

∫ xj (t)xk (t)dt = 0 where j ≠ k

t1

t2

2
Let ∫ x (t)dt = kk
k
t1

Let a function f(t), it can be approximated with this orthogonal signal space by adding the
components along mutually orthogonal signals i.e.

f (t) = C1 x1 (t) + C2 x2 (t)+. . . +Cn xn (t) + fe (t)

n
= Σ Cr xr (t)
r=1

n
f (t) = f (t) − Σ Cr xr (t)
r=1

Mean sqaure error 1 t2 2

ε = ∫ [fe (t)] dt
t 2 −t 2 t1

t2 n
1
2
= ∫ [f [t] − ∑ Cr xr (t)] dt
t2 − t2 t1 r=1

The component which minimizes the mean square error can be found by

dε dε dε
= =. . . = = 0
dC1 dC2 dCk

Let us consider dCk

dε
= 0

6/9
 t2
d 1
n 2
[ ∫ [f (t) − Σ Cr xr (t)] dt] = 0
r=1
dCk t2 − t1 t1

All terms that do not contain Ck is zero. i.e. in summation, r=k term remains and all other terms are
zero.

t2 t2

2
∫ −2f (t)xk (t)dt + 2Ck ∫ [x (t)]dt = 0
k
t1 t1

t2
∫ f (t)xk (t)dt
t1
⇒ Ck =
t2 2
int x (t)dt
t1 k

t2

⇒ ∫ f (t)xk (t)dt = Ck Kk
t1

Mean Square Error

The average of square of error function fe(t) is called as mean square error. It is denoted by ε
(epsilon).

1 t2 2
ε = ∫ [fe (t)] dt
t 2 −t 1 t1

1 t2 2
n
= ∫ [fe (t) − Σ Cr xr (t)] dt
t 2 −t 1 t1 r=1

1 t2 t2 t2
2 n 2 2 n
= [∫ [fe (t)]dt + Σ Cr ∫ xr (t)dt − 2Σ Cr ∫ xr (t)f (t)dt
t 2 −t 1 t1 r=1 t1 r=1 t1

You know that

t2 t2
2 2 2
Cr ∫ xr (t)dt = Cr ∫ xr (t)f (d)dt = Cr Kr
t1 t1

7/9
 1 t2 2 n 2 n 2
ε = [∫ [f (t)]dt + Σ Cr Kr − 2Σ Cr Kr ]
t 2 −t 1 t1 r=1 r=1

1 t2 2 n 2
= [∫ [f (t)]dt − Σ Cr Kr ]
t 2 −t 1 t1 r=1

1 t2 2 2 2 2
∴ ε = [∫ [f (t)]dt + (C K1 + C K2 +. . . +Cn Kn )]
t 2 −t 1 t1 1 2

The above equation is used to evaluate the mean square error.

Closed and Complete Set of Orthogonal Functions

Let us consider a set of n mutually orthogonal functions x1(t), x2(t)...xn(t) over the interval t1 to t2.
This is called as closed and complete set when there exist no function f(t) satisfying the condition
t2
∫ f (t)xk (t)dt = 0
t1

If this function is satisfying the equation then f(t) is said to

t2
∫ f (t)xk (t)dt = 0 for k = 1, 2, . .
t1

be orthogonal to each and every function of orthogonal set. This set is incomplete without f(t). It
becomes closed and complete set when f(t) is included.

f(t) can be approximated with this orthogonal set by adding the components along mutually
orthogonal signals i.e.

f (t) = C1 x1 (t) + C2 x2 (t)+. . . +Cn xn (t) + fe (t)

If the infinite series C1 x1 (t) + C2 x2 (t)+. . . +Cn xn (t) converges to f(t) then mean square error
is zero.

Orthogonality in Complex Functions

If f1(t) and f2(t) are two complex functions, then f1(t) can be expressed in terms of f2(t) as

f1 (t) = C12 f2 (t) ..with negligible error

t ∗
2
∫ f1 (t)f2 (t)dt

Where
t
1
C12 = t
2 2
∫ |f2 (t)| dt
t
1

Where f
∗
2
(t) = complex conjugate of f2(t).

8/9
If f1(t) and f2(t) are orthogonal then C12 = 0

t2
∗
∫ f1 (t)f (t)dt
t1 2
= 0
t2 2
∫ |f2 (t)| dt
t1

t2

∗
⇒ ∫ f1 (t)f (dt) = 0
2
t1

The above equation represents orthogonality condition in complex functions.

9/9
UNIT III FOURIER SERIES REPRESENTATION OF SIGNALS:
 UNIT IV SINGULARITY FUNCTIONS
1.0 INTRODUCTION
Singularity functions are discontinuous functions or their derivatives are
discontinuous.

A singularity is a point at which a function does not possess a derivative. In

other words, a singularity function is discontinuous at its singular points.
Hence a function that is described by polynomial in t is thus a singularity
function. The commonly used singularity functions are:

 Step Function,
 Ramp Function, and
 Impulse Function.

At first, we take up the study of a unit-step function.

1.1 UNIT-STEP FUNCTION

The continuous-time unit-step function is defined as:

Above equation defines what a unit-step function. The value of a unit-step

function is one, for values of t > 0, and it is zero, for values of t < 0. It is
undefined at t = 0. The unit-step function has a value between 0 and 1, at t =
0. The value of the unit-step function changes suddenly, at t = 0. Because of
the step change in unit-step function at t= 0, the value of derivative of unit-
step function is infinite at t = 0. In other words, the unit-step function is
discontinuous at t = 0. It can be seen that the derivative of unit-step function
is zero at all instants, except t = 0.
 Fig.1. Plot of unit step function

The unit step function u(t) is represented as shown in Fig. 1. The unit step
function is used widely in network theory and control theory. It can be seen
that the unit step function has a discontinuity at t = 0 and is continuous for all
other values of t .

1.1.1 Reflection operation on the unit-step function

It is easy to visualize how u(-t) would be. This function, u(-t) is reflected
version of u(t) and is illustrated in Fig. 2.

Fig. 2. Plot of u(-t)

Another example using the unit step function is shown in Fig. 3. This
function is called the signum function and it is written as sgn(t).

Fig.3. Plot of signum function

The signum function is often not used in network theory, but it is used in
communication and control theory. It is expressed in terms of unit step
functions as indicated below

(or) sgn(t)= -1+2u(t)

(or) sgn(t)= u(t)-u(-t)

1.1.2 Shifting operation on the unit-step function

A shifted time function is displayed in Fig. 4.

Fig.4. Shifting and Folding operation on step function

The shifted function can be expressed as shown below:

In a shifted unit step function, defined above equation , the step change
occurs at t = τ, whereas the step change occurs at t = 0 for the unit step
function defined by u(t). It can be seen that the shifted unit step function is
obtained by shifting the unit step function to the right by τ seconds. It is seen
from equation that when t > τ , the argument of the shifted unit step function
is positive and then the function has unit value. When the argument of the
shifted unit step function is negative, the function has zero value. It can be
seen that the argument of the shifted unit step function is negative for t < τ.

Synthesis of a signal

It is possible to use singularity functions to generate or synthesize different

signals. An example is shown below to show how a rectangular pulse signal
can be visualized as the combination of two step functions.

Fig. 5. A rectangular pulse function

A rectangular pulse function of unit amplitude, illustrated in Fig. 5, is

obtained as the combination of a unit step function and a shifted step
function with a magnitude of - 1.

Fig.6. Signals used for synthesis of a pulse signal

When the two signals shown in Fig. 6 are added, we get the rectangular pulse
shown in Fig. 5. We get

g(t)=u(t)-u(t- τ)
A system may receive a single rectangular pulse, as shown in Fig. 5. If this
pulse repeats itself after a fixed period, then the resulting signal is a square-
wave periodic signal.

RAMP FUNCTION

Fig.7. A ramp function

The ramp function is illustrated in Fig. 7. It can be defined as follows:

The ramp function has zero value in the range defined by t < 0. When t > 0,
the ramp function increases linearly with time. The unit-step function and the
ramp function are related. We can define the unit-step function, as the
derivative of the ramp function,. Alternatively, we can state that the ramp
function is the integral of the unit-step function.

dr (t )
u (t ) 
dt
t t
r (t )   u(t )dt   1.dt  t.u(t )
 

Fig.8. Shifted ramp, reflected ramp

The effect of operations on the independent variable of the ramp function is
shown by the sketches in Fig. 8. The plots of the shifted ramp function and
the reflected ramp function are displayed.

Fig. 9. Operations on ramp signal

It is possible to shift the ramp function and then reflect it, as shown in Fig. 9.
The ramp function is a signal generated by some electronic circuits. With
additional electronic circuitry, it is possible to generate saw-tooth waveform
displayed in Fig. 10 . Such a signal is used in a cathode-ray oscilloscope
(CRO) as the timing signal. Such a signal is used in a TV also for horizontal
and vertical scanning.

Fig.10. Saw-tooth waveform, used as sweep signal in CROs

IMPULSE FUNCTION

The unit impulse function, designated δ(t), is also called the Dirac delta
function. Its use in network theory, control theory and signal theory is
widespread and it is important because of its properties and the insight it
offers about the network to which it is applied.

Fig.11. Impulse or Dirac delta function

The impulse function is displayed, as shown in Fig. 11. The impulse function
is related to the unit-step function. It is the integral of the impulse function.
t
u (t )    (t )dt


Alternatively, the unit impulse function is defined as the derivative of the

unit step function, as expressed below

du (t )
 (t ) 
dt
The ideal impulse function is represented by a spike at the origin as shown in
Fig. 11.

The impulse has the following properties.

Sifting/Sampling Property

Given a continuous time-function f(t), sampling property is defined as:

This property is also known as the sifting property. Since  (t - a)=0

when t ≠ a, the above integral can be expressed to be:
Sampling property shows how a function can be sampled at a given instant.

Since the impulse function has value only at t = a, the value of f(t)
when t ≠ a is not important. We call this property as the sifting property.

Time Scaling Property

The impulse response is significant since it reveals the nature of the system.
The poles of impulse response are the poles of the system. We can use
convolution integral to obtain the response of a system to any input. To apply
the convolution integral, we make use of the impulse response.

SYNTHESIS OF WAVEFORMS

It is possible to synthesize waveforms using singularity functions. An

example has already been presented, wherein a pulse function has been
generated as the sum of step functions.
UNIT V:

CONVOLUTION AND CORRELATION OF SIGNALS

1 INTRODUCTION

Convolution is a mathematical way of combining two signals to form a third signal. Convolution
is important because it relates the input signal and impulse response of the system to the output
of the system. Correlation is again a mathematical operation that is similar to convolution.
Correlation also uses two signals to form a third signal. It is very widely used in practice,
particularly in communication engineering. Basically, it compares two signals in order to
determine the degree of similarity between them. Radar, Sonar and digital communications uses
correlation of signals very extensively. Correlation may be cross correlation or auto correlation.
When one signal is correlated with itself to form another signal, it is called auto correlation.

2. CONCEPT OF CONVOLUTION

Convolution is a mathematical operation the input - output relationship of an LTI system. It is

most important operation in LTI continuous-time systems. It relates the input and impulse
response of the system to output.

An arbitrary driving function x(t) can be expressed as a continuous sum of impulse functions.
The response y(t) is then given by the continuous sum of responses to various impulse
components. In fact , the convolution integral precisely expresses the response as a continuous
sum of responses to individual impulse components.
 8/27/2015

UNIT 6: MODELING PHYSICAL

SYSTEMS WITH LINEAR
DIFFERENTIAL EQUATIONS

Learning Objectives
2

After this presentation you will be able to:

 Explain what a differential equation is and how it can represent

dynamics in physical systems.
 Identify linear and non-linear differential equations.
 Identify homogeneous and non-homogenous differential
equations.
 Write input/output equations using derivatives and integrals for
electrical and mechanical systems.

1
 8/27/2015

Linear Dynamic Systems

3

Definition
Linear Differential Equation - a linear combination of derivatives
of an unknown function and the unknown function. Derivatives
capture how system variables change with time.

Linear systems - represented with linear differential equations

Solving a differential equation means find a function that changes

with time that satisfies the equation. The result is a function and not
a number. This function describes how a quantity changes with
respect to the independent variable, usually time in a control system.
This can be done using analytic or numerical methods.
lesson8et438a.pptx

Linear Dynamic Systems

4

Example 8-1 Series RL circuit with a current established initially. What does
current do over time?
i(0)  1 A Write KVL equation around the circuit
+ vr(t)  di( t ) 
v R (t )  i(t )  R v L ( t )  L 
 dt 
i(t) R +
vL (t )  vR (t )  0
L vL(t)
 di( t ) 
L   i( t )  R  0
 dt 

To solve - find i(t) that satisfies the above

equation with initial current of 1 A. Can do
analytically or numerically.

lesson8et438a.pptx

2
 8/27/2015

Linear Differential Equations

5

More Complex Differential Equations

Can have higher order d2 d

i( t )  2  i( t )  7  i( t )  0
Derivatives 2, 3, 4… etc dt 2
dt
Above is a 2nd order linear ODE (ordinary differential equation)

d2 d Squared i makes it
Implied function of
i  2  i  i2  0 non-linear
time. i=i(t) dt 2
dt
Above is 2nd order, non-linear ODE

 d2  2nd order, non-linear ODE

 2 v   sin( v)  v  0
Sine of unknown function v(t)
 dt 

When right-hand side (RHS) is 0, equation called homogeneous. Implies no outside stimulation
lesson8et438a.pptx

Physical System Examples

6

The following circuit creates a homogenous differential equation.

Establish
current R=10k

No I i(0)= 5 A
source t>0 t=0
10 mH
5A

Current can not change instantaneously in inductor. Current varies in

time based in the initial value of i(0)= 5A

lesson8et438a.pptx

3
 8/27/2015

Physical System Examples

7

The following circuit creates a non-homogenous differential equation.

Establish
current R=10k Initial current 0

i(-0)= 0 A
t=0
10 mH
5A

Current builds from initial value. Practical example: Energize a relay

coil with the current source. Current source drives the system.

lesson8et438a.pptx

Identify Example Equations

8

Identify which of the following equations as linear/non-linear and

homogeneous/non-homogeneous.

d 2
3. x( t ) x( t ) 0
dt
d
L. i( t ) R. i( t ) 0
2 dt
d
v sin ( v ) 0
dt

d
6. v 2. v V m. sin( t )
dt
t
2
d d 
4. i 2. i 7. i I o. e 0
d t2 dt
lesson8et438a.pptx

4
 8/27/2015

Differential Equations For Control Systems

9

Equations have constant coefficients and are linear.

Single input stimulation r(t)
Single output variable x(t)
General form
dn d2 d
an  n
x ( t )  ...  a 2  2 x ( t )  a1  x ( t )  a1  x ( t )  b0  r ( t )
dt dt dt
Where an.....a2, a1 a0 and b0 are constants

What can r(t) be?

lesson8et438a.pptx

Differential Equations For Control Systems

10

Typical input functions r(t)

Constant Ramp Sinusoid

100sin(t)
-5 3t
-5cos(t+q)
3 -10t
Where 
10 0.2t-3
and q are
constants
u(t)

Unit-step (square wave)

td t
u(t-td) = 1 after td
0 before
lesson8et438a.pptx

5
 8/27/2015

Characteristics of Linear Systems

11

1.) Multiplying by constant is reflected though system.

r(t) x(t) If input r(t) gives output x(t) then,

Linear
K (r(t)) gives K (x(t))
system
K (r(t)) K (x(t)) I/O proportional

2.) Superposition from circuits holds

r1(t) Linear y1(t) If input r1(t) gives y1(t) and input r2(t) gives y2(t) then
system total output y1(t)+y2(t)
r2(t) y2(t)

Total output is the sum of the individual input

responses. From circuits, transients and sine steady-state lesson8et438a.pptx

Dynamic Equations Input/output

relationships
12

Dynamics represented by integrals and derivatives with respect to time

Electrical element: Resistance Electrical Element: Inductance

L Henrys
i(t) R Ohms iL(t)

+ v(t) + vL(t)
Defining Equations
Defining Equations
d
v( t )  R  i ( t ) v L (t)  L  i L (t)
dt
1
i( t )     v( t )
R
t
or 1
L 0
i L (t)   v L () d  i L (0)
1
i(t)  G  v( t ) Where G 
R

lesson8et438a.pptx
Initial current
at t=0

6
 8/27/2015

Dynamic Equations Input/output

13
relationships
Electrical Element: Capacitance All laws from circuit theory
hold for the analysis of
C Farads circuits with dynamic
iC(t) equations . KCL, KVL, mesh
analysis, nodal
+
vC(t) analysis can all be
performed. Substitute the
Defining Equations appropriate integral or
d Initial voltage derivative into the mesh or
iC (t)  C  vC (t)
dt at t=0 nodal formulation.

t
1
C 0
vC (t)  i C () d  v C (0)

lesson8et438a.pptx

Dynamic Electrical Equations

14

When the current or voltage in a circuit element involves two currents or voltages in
the derivative or integral, take the difference of the voltages or currents

Example 8-2: Write mesh equations for the circuit below using the lumped circuit
element representations

0.1 H Loop 1

0.5 H  v(t )  vL (t )  vR1 (t )  0

i1(t) R2
5 i2(t) 10 vL (t )  vR1 (t )  v(t )
v(t)
d
R1 L i L ( t )  R1  i1 (t )  i 2 (t )
dt
d
0.1 i1 ( t )  5  i1 (t )  i 2 (t )  v(t )
dt
lesson8et438a.pptx

7
 8/27/2015

Example 8-2 Solution (1)

15

0.1 H

0.5 H
i2(t) R2
i1(t)
5 10
v(t)
R1 Loop 2
vR1 (t )  vC (t )  vR 2 (t )  0
t
1
C 0
 i 2 () d  v C (0)

t
1
5  i 2 (t )  i1 (t )   i 2 () d  v C (0)  10  i 2 (t )  0
0.5 0

lesson8et438a.pptx

Example 8-2 Solution (2)

16

Mesh equations form a system of integral-differential

equations in unknown functionsi1(t) and i2(t).

5  i1 ( t )  i 2 ( t )   0.1
d
Loop 1 i1 ( t )  v( t ) (1)
dt
1/0.5=2
t
Loop 2 5  i 2 ( t )  i1 ( t )   2   i 2 () d  v C (0)  10  i 2 ( t )  0 (2)
0

Solution techniques: convert all equations into derivatives only. Approximate derivatives
using mathematical methods and calculate approximate derivative values for some small
increment in time. Results are a list of computed points that approximate variable over a
time interval. Graph these points to see system response
lesson8et438a.pptx

8
 8/27/2015

Example 8-3 OP AMP Differentiators

17

Use rules of circuit analysis and ideal OP AMPs to find the input/output
relationship for the circuit below.
Rules of OP AMPS
1.) No current flows into OP AMP
if 2.) V- = V+
v-
iin
Use nodal analysis at OP AMP inverting node.
v+
Sum currents at inverting input
iin(t)+if(t)=0 so iin(t)=-if(t)
iin(t)=iC(t)
Define ic(t) in terms of voltage Feedback current
d
iC (t)  C  vC (t) vo (t )  v  (t )
dt if (t ) 
Rf
v C ( t )  v in ( t )  v  ( t )
lesson8et438a.pptx

Example 8-3 Solution (2)

18

Complete derivation
i in ( t )  i f ( t )
 v (t)  v (t) 
C
d
 
v in ( t )  v  ( t )    o 
V+(t)=V-(t)=0
Positive terminal
dt  Rf  grounded
  v (t) 
C  v in ( t )   o 
d
dt  Rf 
 Rf C
d
vin ( t )  v o ( t )
dt

Input Output
Circuit takes derivative of input voltage
lesson8et438a.pptx

9
 8/27/2015

Mechanical System Models

19

Self-regulating tank system

Need relationship for how level changes
with time. Derive differential equation.

V  (Qin  Qout )  t 1

Write level change in terms of tank volume

V
h  2
A 2

Qin>Qout h increases V (Qin  Qout )  t

h   1 2
A A
Qin<Qout h decreases

lesson8et438a.pptx

Mechanical System Models

20

Self-regulating tank system – Find the average level change over

time interval t.
V (Qin  Q out )  t
h  
t A A t

h (Qin  Q out )

t A

Take limit as time h dh (Qin  Qout ) Write Qout in terms of

lim   system parameters
interval goes to zero t 0 t dt A

lesson8et438a.pptx

10
 8/27/2015

Self-Regulating Tank
21

Assume laminar flow for simplicity. Qout determined by pressure at bottom of

tank.
p  R L  Q out Combine these two
equations and solve
p  g h for Qout
p g h
Q out  
RL RL

Bring all terms with g h

Qin 
h or derivative of h dh RL dh Qin   g  h
   
to one side dt A dt A RL A

 R L  A  dh  R L  A     g   Q  R  A 
          h   in  L 
   g  dt    g   L R  A   A    g 
lesson8et438a.pptx

Self-Regulating Tank
22

Simplified result
 R L  A  dh R 
    h  Qin  L 
   g  dt  g 

Qin is independent of the liquid height. consider it along with density, area
and flow resistance to be constant

 R A  R 
Let    L  G   L 
 g   g 

Non-homogeneous equation that determines how height of

dh
   h  G  Qin liquid in tank varies with time.
dt Shutting off Qin finds homogeneous response of tank height

lesson8et438a.pptx

11
 8/27/2015

Mechanical Models
23

Non-Self-Regulating Tank (Pumped Drainage)

Output flow is fixed by pump flow rate. It

is independent of the liquid height in tank.

V (Qin  Qout )  t
h  
A A
Define: t = t1 - t0
h(t) = h(t1) - h(t0)
h (Qin  Qout ) h dh
 lim 
t A t 0 t dt

lesson8et438a.pptx

Mechanical Models
24

Non-Self-Regulating Tank (Pumped Drainage)

dh (Qin  Qout )

dt A Independent
of h

dh To solve this differential equation, integrate

both sides of the equation with respect to t.
t1 t1

 dt dt   A Q  Qout  dt
dh 1
in
t0 t0

t1

h(t1 )  h ( t 0 )  
1
Qin  Qout  dt
t0
A
Note: Right hand side is not a function of t. t
1
It is all a constant and can be taken out of h(t) 
1
Qin  Qout  dt
the integral. A t0

lesson8et438a.pptx

12
 8/27/2015

Non-Self-Regulating Tank (Pumped Drainage)

25

If Qin is not a constant but changes with time, the formula below models the
response of the tank system
t1

h(t1 )  h ( t 0 )  
1
Qin (t )  Qout  dt
t0
A Definite integral
t1
from calculus
h(t1 )   Qin ( t )  Q out  dt  h ( t 0 )
1
t0
A

Initial tank height

Final tank height

For pumped tank with constant input and output flows, tank drains linearly with
time based on the difference between the flow rates. Final tank height depends
on the pump flow rate and the time the pump operates.
lesson8et438a.pptx

Thermal Systems
26

Heating characteristic of a liquid filled thermometer

Ta = fluid temperature
Ta Tm = measured temperature
CTm = thermal capacitance of thermometer
Tm
How does measured temperature change with time? Heat
transferred to thermometer depends on T, RT and time
CTm interval

Q 
Ta  Tm  t Incremental conduction heat flow
RT

Definition of thermal Q Q
 CTm   Tm
capacitance Tm CTm

lesson8et438a.pptx

13
 8/27/2015

Thermal Systems
27

Derive the thermal equation

Q Ta  Tm   t
 Divide both sides by CTm
CTm R T  CTm

Tm 
Ta  Tm  t
Use definition of CTm from last slide
R T  CTm

Determine the average change in temperature for a t

Tm Ta  Tm 

t R T  CTm

Tm dTm Ta  Tm 

Take limit t approaches 0 lim  
t 0 t dt R T  CTm

lesson8et438a.pptx

Thermal Systems
28

Get all Tm and its derivatives on one side of the equation

dTm Ta  Tm 

dt R T  CTm
dTm Ta Tm
 
dt R T  CTm R T  CTm

dTm Tm Ta
  Move Tm to same side as derivative of Tm
dt R T  CTm R T  CTm

 R T  CTm    R T  CTm 
dTm Tm Ta
R T  CTm  
dt R T  CTm R T  CTm

dTm This is similar in form to the self-regulating tank equation.

R T  CTm   Tm  Ta
dt This is a non-homogeneous differential equation that
describes how the measured temperature changes with time
lesson8et438a.pptx

14
 8/27/2015

Mechanical Systems
29

Pneumatic Control Value Position Determine how the position of a air

actuated control value changes with time
after pressure is applied.
1/K=Cm
Fa
Free body diagram- All forces must sum to zero
FCm(t)
Fa Fa = Pa(A) = input air force
FI(t)
x=0 v All forces must balance at
each instance in time so:
M
FI(t) = inertial force
B=Rm
FCm(t) = spring force
FRm(t) = viscous friction force
FRm(t)
lesson8et438a.pptx

Mechanical Systems- Control Valves

30

All forces must sum to zero- Assume down is positive direction

Fa  FRm ( t )  FI ( t )  FCm ( t )  0
FCm(t)
Fa
Fa  FRm ( t )  FI ( t )  FCm ( t )
FI(t)
Need equation that relates position, x, to time

M Friction force
dx(t)
FRm ( t )  R m  v( t ) Remember v(t) 
dt
dx(t)
FRm(t) FRm ( t )  R m  Viscous friction is proportional
dt
to velocity. Velocity = rate of
change of position
lesson8et438a.pptx

15
 8/27/2015

Mechanical Systems- Control Valves

31

Spring force
 1 
FCm ( t )     x ( t )  K  x ( t )
 Cm 

Force from spring is proportional to its length, x. Cm is spring capacitance,

K = spring constant. So 1/K=Cm.

Inertial force
d  dx(t)  d 2 x(t)
a(t)   
FI ( t )  M  a ( t ) dt  dt  dt 2
dv(t) dx(t)
a(t)  v(t)  d 2 x(t)
dt dt FI ( t )  M  a ( t )2
dt
lesson8et438a.pptx

Control Valve Model

32

Combine individual terms

Fa  FRm (t )  FI (t )  FCm (t )

dx(t) d 2 x(t)  1 
Fa  R m   M    x(t)
dt dt 2  Cm 

Fa is constant. Equation describes how position changes with time. Second

order equation, non-homogeneous.

lesson8et438a.pptx

16
 8/27/2015

Summary of Mechanical Models

33

Self-Regulating Tank Heat Transfer

dh dTm
  h  G  Qin R T  CTm   Tm  Ta
dt dt
 R A  R 
   L  G   L 
 g   g  Control Valve Position
d 2 x(t) dx(t)  1 
Non-Self-Regulating Tank Fa  M  2
 Rm      x ( t )
(Pumped Drainage) dt dt  Cm 

t1

h(t1 )  
1
Qin (t )  Qout  dt  h(t 0 )
t0
A
lesson8et438a.pptx

34 ET 438a Automatic Control Systems Technology

End Lesson 8: Modeling Physical Systems

With Linear Differential Equations

lesson8et438a.pptx

17
UNIT 7

FOURIER TRANSFORMS
AND SAMPLING :
SAMPLING:
UNIT 8

LAPLACE TRANSFORMS
UNIT 9
 The z-Transform

The z-Transform is defined as:

∞
X (z) := ∑ x [ n ] z−n (1)
n=−∞

where z ∈ C. It recovers the DTFT when z is on the unit circle, i.e.,

when z = e jω for some ω ∈ [0, 2π ):

X (z) = X (e jω ). (2)
z=e jω

Thus, the DTFT converges if the region of convergence (ROC) for the
z-transform includes the unit circle.

Im
ejω

ω
z-­‐Transform  developed  in  the  1Re950s:  

Example 1: x [n] = δ[n] → X (z) = 1, ROC: entire complex plane

Example 2: x [n] = an u[n]
∞ ∞
1
X (z) = ∑ an z−n = ∑ (az−1 )n = 1 − az−1 if | az−1 | < 1
n =0 n =0
| {z }
ROC: |z|>| a|
z
Poles and zeros: X (z) = z− a → pole at z = a, zero at z = 0.
Yakov  Tsypkin  
DTFT: John  Ragazzini   Eli  Jury   Lo�i  Zadeh  
1
       (Moscow)   X ( e jω
)      (1Columbia)  
=
− ae− jω
if | a| < 1

Prof. Zadeh: Berkeley EECS Profs

Im unit
Dear  Murat,  
circle
[…]  I  am  pleased  to  learn  that  you  are  talking  about  z-­‐transforma�on  in  your  lecture  on    
Monday.  […]  My  paper  with  Ragazzini  was  published  in  1952.  Ragazzini  was  my  Ph.D.    
supervisor.  […]  In  the  period  before  athe  paper  
Re was  published  there  was  a  growing  recogni�on    
that  sampling  a  signal  was  an  important  part  of  signal  processing.  Sampling  
Figure 1: Formerplayed   a  key  
Berkeley EECS role  
pro-
in  Shannon's  work.  At  that  �me,  it  occurred  
ROC to  me  that  it  was  natural  to  look  at  sampling  as  a  
fessors Lotfi Zadeh (above) and Eliahu
Jury (below) were among those who
convolu�on  of  a  signal  with  a  sequence  of  delta-­‐func�on,  transforming   the  the
developed signal  
theoryinto   what    
of z transforms
was  called  the  z-­‐transform.  I  used  the  le�er  z  not  because  of  my  name   but  because  in  the    
in the 1950s. Research in sampling was
partly motivated by radar which came
mathema�cal  literature  on  difference  equa�ons,  z  was  the  symbol  that  was  used.  
to prominence during[World
…]   War II.
Sincerely,  Lo�i  
Example 3: x [n] = − an u[−n − 1]

−1 ∞ Im
X (z) = ∑ − an z−n = ∑ − a−n zn
n=−∞ n =1
∞
= 1− ∑ ( a −1 z ) n
n =0
1
a Re
= 1− if | a−1 z| < 1
1 − a −1 z
− a −1 z 1 ROC
= = ROC : |z| < | a|
1 − a −1 z 1 − az−1

DTFT converges is | a| > 1.

 n  n
Example 4: x [n] = − 1
u[−n − 1] + −1 u[n]
2 3

1
1 1 z z 2z(z − 12 )
X (z) = + = + =
1 − 21 z−1 1 + 13 z−1 z− 1
2 z+ 1
3 (z − 1 1
2 )( z + 3 )
1 1
|z| < |z| >
| 2 {z 3}
ROC: 1 <| z |< 1
3 2

Im

− 31 1
2
Re

 n  n
Example 5: x [n] = 1
u[n] − −1 u[−n − 1]
2 3

ROC = {z : |z| > 21 } ∩ {z : |z| < 31 } = ∅

Example 6: x [n] = an , a 6= 0.
x [n] = an u[n] + an u[−n − 1]
ROC = {z : |z| > a} ∩ {z : |z| < a} = ∅

Properties of the ROC

Section 10.2 in Oppenheim & Willsky
As seen in the examples above, the ROC is a ring or disk in the z-
plane, centered at the origin. Note that it does not contain any poles.
If x [n] is right-sided (e.g., Example 1) then the ROC extends from the
outermost pole to ∞.
Inverse z-Transform by Partial Fraction Expansion

Example 7:

1 A1 A2
X (z) =   = 1 −1
+
1 − 21 z−1

1 − 41 z−1 1 − 12 z−1 1 − 4z

 
1 −1
A1 = 1− z X (z)|z= 1 = −1
4 4
 
1
A2 = 1 − z−1 X (z)|z= 1 = 2
2 2

1) 2)

− 41 1
2

ROC
  n  n    n  n 
1 1 1 1
x [n] = 2 − u[n] x [n] = − 2 − u[−n − 1]
2 4 2 4
3)

 n  n
1 1
x [ n ] = −2 u[−n − 1] − u[n]
2 4

How to perform a PFE in general?

b0 + b1 z−1 + . . . + b M z− M
X (z) = , a0 6 = 0
a 0 + a 1 z −1 + . . . + a N z − N

Suppose unrepeated poles: d1 , d2 , ..., d N .

If M < N,
N
Ak
X (z) = ∑ 1 − d k z −1
k =1
If M ≥ N,
M− N N
Ak
X (z) = ∑ Br z−r + ∑ 1 − d k z −1
r =0 k =1
⇓
M− N N
x [n] = ∑ Br δ[n − r ] + ∑ Ak dnk u[n]
r =0 k =1
| {z }
if right-sided

Example 8:

1 + 2z−1 + z−2
X (z) = M=N=2
1 − 32 z−1 + 21 z−2
A1 A2
= B0 + +
1 −1
1 − 2z 1 − z −1

Matching coefficients: A1 = −9, A2 = 8, B0 = 2.

 n
1
x [n] = 2δ[n] − 9 u[n] + 8u[n].
2

Differentiation (in z-domain) Property:

Z
←→ X (z)
x [n] ROC = R
Z dX (z)
nx [n] ←→ −z ROC = R
dz
∞
Proof: X (z) = ∑ x [ z ] z−n
n=−∞
∞ ∞
dX (z)
dz
= ∑ −nx [n]z−(n+1) = −z−1 ∑ nx [n]z−n
n=−∞ n=−∞
∞
dX (z)
∑ nx [n]z−n = −z
dz
n=−∞

Example 9:

Z 1
an u[n] ←→
1 − az−1
az−2 az−1
 
Z d 1
nan u[n] ←→ −z = z =
dz 1 − az−1 (1 − az−1 )2 (1 − az−1 )2

Back to Partial Fraction Expansions: If dk is a pole of multiplicity

two, include two terms:
1 d k z −1
Ak1 −
+ Ak2
1 − dk z 1 (1 − d k z −1 )2
l
( Ak1 + Ak2 n) dnk u[n]
Example 10: X (z) − 12 + z−1 M=1 1
= |z| >
(1 − 21 z−1 )2 N=2 2
1 −1
1 2z
= A11 + A12  2
1 − 21 z−1 1 − 12 z−1
A11 + 12 ( A12 − A11 ) z−1
=  2
1 − 12 z−1
)
A11 = − 21 3
1 A12 =
2 ( A12 − A11 ) = 1
2
   n
1 3 1
x [n] = − + n u[n]
2 2 2

Signal Transform ROC Table 1: z transforms of several func-

tions.
δ[n] 1 all z
δ[n − m] z−m all z except z = 0 if m > 0,
all z except z = ∞ if m < 0
1
u[n] 1 − z −1
|z| > 1
1
−u[−n − 1] 1 − z −1
|z| < 1
1
an u[n] 1− az−1
|z| > a
1
− an u[−n − 1] 1− az−1
|z| < a
az−1
nan u[n] (1− az−1 )2
|z| > a
az−1
−nan u[−n − 1] (1− az−1 )2
|z| < a
1−cos(ω0 )z−1
cos(ω0 n)u[n] 1−2 cos(ω0 )z−1 +z−2
|z| > 1
sin(ω0 )z−1
sin(ω0 n)u[n] 1−2 cos(ω0 )z−1 +z−2
|z| > 1
1−r cos(ω0 )z−1
r n cos(ω0 n)u[n] 1−2r cos(ω0 )z−1 +r2 z−2
|z| > r
r sin(ω0 )z−1
r n sin(ω0 n)u[n] 1−2r cos(ω0 )z−1 +r2 z−2
|z| > r

Properties of the z-Transform

1) Linearity: ax1 [n] + bx2 [n] ←→ aX1 (z) + bX2 (z)

ROC contains R1 ∩ R2 where Ri is the ROC of xi [n], i = 1, 2.
2) Time Shifting: x [n − n0 ] ←→ z−n0 X (z)
ROC unchanged, except for possible addition/deletion of 0 and ∞.
Example 11: Find the inverse z-Transform (right-sided) of

1 1
X (z) =  =z
1 − 12 z−1

1 −1
z −1
1− 2z
  n +1
1
x [n] = u [ n + 1]
2

3) Scaling in the z-domain:

 
Z z
z0n x [n] ←→ X ROC = |z0 | · R
z0

where R is the ROC of x [n]. Compare to:

 
DTFT
e jω0 n x [n] ←→ X e j(ω −ω0 )
L
e s0 t x ( t ) ←→ X (s − s0 ) ROC = R + Re{s0 }

4) Time Reversal:
 
Z 1
x [−n] ←→ X ROC = 1/R
z

Example 12:
 n
1 1 1
x [n] = u[n] ↔ X (z) = 1 −1
|z| >
2 1− 2z
2
−2z−1
 
1 1
x [−n] = 2n u[−n] ↔ X = = |z| < 2
z 1 − 12 z 1 − 2z−1

5) Convolution Property:

Z
x1 [n] ∗ x2 [n] ←→ X1 (z) X2 (z) ROC contains R1 ∩ R2

6) Differentiation in z-domain:

Z dX (z)
nx [n] ←→ −z ROC unchanged
dz
Proof and example on page 4.
7) Initial Value Theorem: If x [n] = 0 for n < 0, then

x [0] = lim X (z)

z→∞

Proof: X (z) = x [0] + x [1]z−1 + x [2]z−2 + ...

| {z }
→0 as z→∞