11/18/2021

EEE 3105: Signals and Linear Systems

2.4.2 Tabular Convolution Method

Example 2.4.2.A: Convolute x1[n]  x2 [n] by Tabular Method, where
x1[n]  {1, 2,  2} and x2 [n]  {2, 0, 1}
 

Solution:

1
 11/18/2021

Example 2.4.2.B: Convolute x1[n]  x2 [n] by Tabular Method, where

x1[n]  {1, 2, 0, 1} and x2 [n]  {3, 1, 0, - 1}
 

Solution:

Problem: Convolute x1[n]  x2 [n], where x1[n]  {1, - 1, 1} and



x2 [n]  {2, - 1, 1}

3

2.4.3 Column/Shortcut Convolution Method

Example 2.4.3.A: Convolute x[n]  h[n] by Shortcut Method, where

x[n]  {0, 1, 2, 1, 0} and h[n]  {2, 0, 0, - 1}
 

Solution:
0 1 2 1 0
* 2 0 0 -1 Here, x[n] = 0, 1, 2, 1, 0; 0  n  4
h[n] = 2, 0, 0, -1; 0  n  3
0 2 4 2 0
So, output range  [(0) + (0)]  n  [(4)+(3)]
0 0 0 0 0  0  n 7
Duration of convolved output being [(5+4)-1]
0 0 0 0 0
= 8 samples
0 -1 -2 -1 0
y[n] = {0, 2, 4, 2, -1, -2, -1, 0}

2
 11/18/2021

Convolute x[n]  h[n] by Shortcut Method, where

Example 2.4.3.B:
x[n]  {0, 2, 2, 2, 0} and h[n]  {0, 0, 1, 2}
 

Solution:
0 2 2 2 0
* 0 0 1 2
Here, x[n] = 0, 2, 2, 2, 0; -2  n  2
0 0 0 0 0
h[n] = 0, 0, 1, 2; -2  n  1
0 0 0 0 0 So, output range  [(-2) + (-2)]  n  [(2)+(1)]
 -4  n  3
0 2 2 2 0
Duration of convolved output being [(5+4)-1] =
0 4 4 4 0 8 samples

y[n] = {0, 0, 0, 2, 6, 6, 4, 0}

2.5 Properties of Linear Time-Invariant Systems

The extremely important formulae/representations of continuous-time an
discreate-time LTI systems in terms of their unit impulse response are:

y[n]   x[k ]h[n  k ]  x[n] * h[n] (2.5.1)
k  

y (t )    x( )h(t  )d  x(t ) * h(t ) (2.5.2)
One consequence of these representations is that the characteristics of an LTI
system are completely determined by its impulse response. It is important to
emphasize that these property holds in general only for LTI systems. The
Properties of Linear Time-Invariant Systems are as follows:

(i) Commutative property, (ii) Distributive property,

(iii) Associative property, (iv) LTI systems with and without memory,
(vi) Invertibility of LTI systems, (vii) Causality of LTI systems,
(viii) Stability of LTI systems, (ix) The unit step response of an LTI system
6

3
 11/18/2021

(i) Commutative Property

The basic property of convolution in both discrete and continuous-time
is that it is a commutative operation. That is, in discrete-time

x[n] * h[n]  h[n] * x[n]   h[k ]x[n  k ] (2.5.3)
k 

and in continuous-time

x(t ) * h(t )  h(t ) * x(t )   h( ) x(t  )d (2.5.4)
These expressions can be verified in a straightforward manner. For
example, in the discrete-time case, let r = n-k,  k = n-r, Eq.(2.5.3)
becomes
 
x[n] * h[n]   x[k ]h[n  k ]   x[n  r ]h[r ]  h[n] * x[n] (2.5.5)
k  k 

Fig.2.5.1

(ii) Distributive Property

Another basic property of convolution is distributive property.
Specifically, convolute distributes over addition, so that in
discrete-time
x[n] * h1[n]  h2 [n]  x[n] * h1[n]  x[n] * h2 [n] (2.5.6)

and in continuous-time
x(t ) * [h1 (t )  h2 (t )]  x(t ) * h1 (t )  x(t ) * h2 (t ) (2.5.7)

Fig.2.5.2

4
 11/18/2021

(iii) Associative Property

Another impotent and useful property of convolution is that it is
associative. That is, in discrete-time

x[n] * h1[n] * h2 [n]   x[n] * h1[n] * h2 [n] (2.5.8)

and in continuous-time
x(t ) * h1 (t )  h2 (t )  x(t ) * h1 (t )* h2 (t ) (2.5.9)

Fig.2.5.3

(iv) Linear Systems with and without Memory

A system is memoryless when its output at any time depends only on the
value of the input at that same time. From Eq.(2.5.1), for discrete-time
LTI system this can only be true if h[n] = 0 for n  0. In this case the


impulse response has the form y[n]   x[k ]h[n  k ]  x[n] * h[n]
k  

h[n]  K [n] (2.5.10) Though the amplitude of [n] is unity,

but its response may or may not be unity.

where, K = h[0] is constant, and the convolution sum reduces to the

relation
y[n]  Kx[n] (2.5.11)
Similarly, for continuous-time case, the relation becomes
y(t )  Kx(t ) (2.5.12)
For some constant K and has the impulse response
h(t )  K (t ) (2.5.13)
If K = 1 then, Eq.(2.5.10) and Eq.(2.5.11) become identity systems.
If the condition, h[n] = 0 for n  0 is not satisfied then the system will
be with memory. 10

5
 11/18/2021

(v) Invertibility of LTI systems

A continuous-time LTI system with impulse response h(t) and an inverse
system having impulse response h1(t) connected in series is shown in
Fig.2.5.4(a). The system is invertible only when it produces an output equal
to the input of the first system. The inverse system with impulse response
h1(t), results in w(t) = x(t)  such that series interconnection in Fig.2.5.4(a)
is identical to the identity system in Fig.2.5.4(b). Since the overall impulse
response in Fig.2.5.4(a) is h(t)*h1(t), we have the condition that h1(t) must
satisfy for it to be the impulse response of the inverse system, namely,
h(t ) * h1 (t )   (t ) (2.5.14)
Similarly, in discrete-time, the impulse response h1[n] of the inverse system
for an LTI system with impulse response h[t] must satisfy
h[n] * h1[n]   [n] (2.5.15)
Fig.2.5.4 Concept of an
inverse system for continuous-
time LTI systems. The system
with impulse response h1(t) is
the inverse of the system with
impulse response h(t) if
h(t)*h1(t) = (t)
11

(vi) Causality for LTI systems

The output of a causal system depends only on the present and past
values of the input to the system. In order for a discrete-time LTI
system to be causal, y[n] must not depend on x[k] for k > n.
According to the Eq. (2.5.1), it is to be true that all of the coefficients
h[n- k] that multiply values of x[k] for k > n must be zero. So, the
impulse response of a causal discrete-time LTI system must satisfy
the condition
h[n]  0 for n  0 (2.5.16)
Explanations: We know,

y[n]   x[k ]h[n  k ]
k  

Let, n = 1, the above equation becomes


y[1]   x[k ]h[1  k ]
k  
12

6
 11/18/2021

If k > n, let, k = 2, then the equation becomes

 
y[1]   x[2]h[1  2]   x[2]h[1]
k   k  

Here, x[2] shows the future value. So, the system is NOT CAUSAL.

This system will be CAUSAL only when h[n-k] or h[n] = 0 when n < 0,
which makes all the FUTURE terms ZERO.

Example: Let k ranges from 0 to 2. So, the general equation can be

expanded as
y[n]  x[0]h[n  0]  x[1]h[n  1]  x[2]h[n  2]

When n = 1, then this equation becomes

y[1]  x[0]h[1  0]  x[1]h[1  1]  x[2]h[1  2]
 x[0]h[1]  x[1]h[0]  x[2]h[1]
13

According to the causality condition we can write this equation as

y[1]  x[0]h[1]  x[1]h[0]  x[2]  0  x[0]h[1]  x[1]h[0]  0

So, for causal discrete-time LTI system, the condition in Eq.(2.5.16)

implies that the convolution sum representing in Eq.(2.5.1) becomes
n
y[n]   x[k ]h[n  k ] (2.5.17)
k  

and the alternative equivalent form, Eq.(2.5.3), becomes


y[n]   h[k ]x[n  k ] (2.5.18)
k 0

Similarly, a continuous-time LTI system is causal if

h(n)  0 for t  0 (2.5.19)
In this case the convolution integral is given by

y[n]    x( )h(t  )d  0 h( ) x(t  )d
t
(2.5.20)
14

7
 11/18/2021

n 

(vii) Stability for LTI systems y[n]   x[k ]h[n  k ]   h[k ]x[n  k ]
k   k  

A system is stable if every bounded input produces a bounded output. In

order to determine conditions under which LTI systems are stable, consider
an input x[n] that is bounded in magnitude:
x[n]  B for all n (2.5.21)
Suppose that we apply this input to an LTI system with unit impulse response
h[n]. Then, using the convolution sum, we obtain an expression for the
magnitude of the output:

y[n]   h[k ]x[n  k ] (2.5.22)
k  
Since the magnitude of the sum of a set of numbers is no larger than the sum
of the magnitudes of the numbers, it follows from Eq.(2.5.22) that

y[n]   h[k ] x[n  k ] (2.5.23)
k  
From Eq.(2.5.21), x[n-k]< B for all values of k and n. Together with Eq.
(2.5.23), this implies that

y[n]  B  h[k ] for all n (2.5.24)
k   15

From eq. (2.5.24), we can conclude that if the impulse response is

absolutely summable, that is, if

 h[k ]   (2.5.24)
k  

then y[n] is bounded in magnitude, and hence, the system is stable.

In continuous time, we obtain an analogous characterization of stability

in terms of the impulse response of an LTI system. Specifically, if
x(t)< B for all t, then, in analogy with Eq.(2.5.22)-(2.5.24), it follows
that

y (t )   h( ) x(t  )d

  h( ) x(t   ) d

 B  h( ) d
Therefore, the system is stable if the impulse response is absolutely
integrable, i.e., if

  h( ) d   (2.5.25)
16

8
 11/18/2021

(viii) The Unit Step Response of an LTI systems

There is another signal that is also used quite often in describing the
behavior of LTI systems: the unit step response, s[n] or s(t),
corresponding to the output when x[n] = u[n] or x(t) = u(t). We will find
it useful on occasion to refer to the step response, and therefore, it is
worthwhile relating it to the impulse response. From the convolution –
sum representation, the step response of a discrete-time LTI system is the
convolution of the unit step with the impulse response; that is,
s[n]  u[n] * h[n]
However, by the commutative property of convolution,

s[n]  h[n] * u[n]   h[k ]u[n  k ] (2.5.26)
K  
and therefore, s[n] can be viewed as the response to the input h[n] of a
discrete-time LTI system with unit impulse response u[n]. Since u[n-k]
is 0 for [n-k] < 0 and 1 for [n-k]  0, so Eq.(2.5.26) becomes an
accumulator with unit impulse response u[n],
n
s[n]   h[k ] (2.5.27)
k   17

The relationship between the input and output of an accumulator can be

described as
n 1
s[n]   h[k ]  h[n]
k  

or, s[n]  s[n  1]  h[n] (2.5.28)

From this equation, it is clear that h[n] can be recovered from s[n]
using the relation
h[n]  s[n]  s[n  1] (2.5.29)
As in Eq.(2.5.27), the step response of a discrete-time LTI system is the
running sum of its impulse response. Conversely, the impulse
response of a discrete-time LTI system is the first difference of its step
response [Eq.(2.5.29)].

Similarly, for continuous-time we can write

s(t )  - h[ ]d
t
(2.5.30)
18

9
 11/18/2021

The unit impulse response can be determined as the first derivative of

the unit step response,

ds(t )
h(t )   s(t ) (2.5.31)
dt
Therefore, in both continuous and discrete time, the unit step response
can also be used to characterize an LTI system, since we can calculate
the unit impulse response from unit step response .

19

10