ELG 3120 Signals and Systems Chapter 1

Chapter 1 Signal and Systems

1.1 Continuous-time and discrete-time Signals

1.1.1 Examples and Mathematical representation

Signals are represented mathematically as functions of one or more independent variables. Here
we focus attention on signals involving a single independent variable. For convenience, this will
generally refer to the independent variable as time.

There are two types of signals: continuous-time signals and discrete-time signals.

Continuous-time signal: the variable of time is continuous. A speech signal as a function of

time is a continuous-time signal.

Discrete-time signal: the variable of time is discrete. The weekly Dow Jones stock market index
is an example of discrete-time signal.
x(t) x[n]

x[0]

x[-1] x[1]
x[-2] x[2]
-5 -4 -3 3 4 5
n
t -2 -1 0 1 2

Fig. 1.1 Graphical representation of continuous- Fig. 1.2 Graphical representation of discrete-time
time signal. signal.

To distinguish between continuous-time and discrete-time signals we use symbol t to denote the
continuous variable and n to denote the discrete-time variable. And for continuous-time signals
we will enclose the independent variable in parentheses (•), for discrete-time signals we will
enclose the independent variable in bracket [•].

A discrete-time signal x[n] may represent a phenomenon for which the independent variable is
inherently discrete. A discrete-time signal x[n] may represent successive samples of an
underlying phenomenon for which the independent variable is continuous. For example, the
processing of speech on a digital computer requires the use of a discrete time sequence
representing the values of the continuous-time speech signal at discrete points of time.

1/1 Yao
 ELG 3120 Signals and Systems Chapter 1

1.1.2 Signal Energy and Power

If v(t ) and i (t ) are respectively the voltage and current across a resistor with resistance R , then
the instantaneous power is

1 2
p(t ) = v(t )i(t ) = v (t ) . (1.1)
R

The total energy expended over the time interval t1 ≤ t ≤ t 2 is

1 2
∫ p(t )dt = ∫
t2 t2
v (t )dt , (1.2)
t1 t1 R

and the average power over this time interval is

1 1 1 2
∫ ∫
t2 t2
p (t )dt = v (t )dt . (1.3)
t 2 − t1 t1 t 2 − t1 t1 R

For any continuous-time signal x(t ) or any discrete-time signal x[n] , the total energy over the
time interval t1 ≤ t ≤ t 2 in a continuous-time signal x(t ) is defined as

t2 2
∫t1
x (t ) dt , (1.4)

where x denotes the magnitude of the (possibly complex) number x . The time-averaged power
1 2
∫
t2
is x (t ) dt . Similarly the total energy in a discrete-time signal x[n] over the time
t 2 − t1 t1
interval n1 ≤ n ≤ n2 is defined as

n2

∑ x[n]
2
(1.5)
n1

1 n2

∑
2
The average power is x[n ]
n2 − n1 + 1 n1

In many systems, we will be interested in examining the power and energy in signals over an
infinite time interval, that is, for − ∞ ≤ t ≤ +∞ or − ∞ ≤ n ≤ +∞ . The total energy in continuous
time is then defined

T 2 ∞ 2
E ∞ = lim ∫ x(t ) dt = ∫ x(t ) dt , (1.6)
T → ∞ −T −∞

2/1 Yao
 ELG 3120 Signals and Systems Chapter 1

and in discrete time

+N +∞
E ∞ = lim ∑ x[n] = ∑ x[n ] .
2 2
(1.7)
N →∞
−N −∞

For some signals, the integral in Eq. (1.6) or sum in Eq. (1.7) might not converge, that is, if x(t )
or x[n] equals a nonzero constant value for all time. Such signals have infinite energy, while
signals with E ∞ < ∞ have finite energy.

The time-averaged power over an infinite interval

1 2

∫
T
P∞ = lim x (t ) dt (1.8)
T → ∞ 2T −T

1 +N
∑
2
P∞ = lim x[n ] (1.9)
N →∞ 2 N + 1
−N

Three classes of signals:

• Class 1: signals with finite total energy, E ∞ < ∞ and zero average power, (Energy Signal)

E∞
P∞ = lim =0 (1.10)
T → ∞ 2T

• Class 2: with finite average power P∞ . If P∞ > 0 , then E ∞ = ∞ . An example is the signal
x[n ] = 4 , it has infinite energy, but has an average power of P∞ =16. (Power Signal)

Class 3: signals for which neither P∞ and E∞ are finite. An example of this signal is x(t ) = t .

1.2 Transformations of the independent variable

In many situations, it is important to consider signals related by a modification of the

independent variable. These modifications will usually lead to reflection, scaling, and shift.

1.2.1 Examples of Transformations of the Independent Variable

3/1 Yao
ELG 3120 Signals and Systems Chapter 1

x[n] x[n-n 0]

n n
n0

(a) (b)

Fig.1.3 Discrete-time signals related by a time shift.

x(t-t0 )
x(t)

t
t0
t

Fig. 1.4 Continuous-time signals related by a time shift.

x[n] x[-n]

n n

(a) (b)

Fig. 1.5 (a) A discrete-time signal x[n] ; (b) its reflection, x[− n] about n = 0 .

x(t) x(-t)

t t
0 0

(a) (b)

Fig. 1.6 (a) A continuous-time signal x(t ) ; (b) its reflection, x(−t ) about t = 0 .

4/1 Yao
 ELG 3120 Signals and Systems Chapter 1

x(t)
x(2t)

t
t 0
(a) (b)
x(t/2)

t
0
(c)

Fig. 1.7 Continuous-time signals related by time scaling.

1.2.2 Periodic Signals

A periodic continuous-time signal x(t ) has the property that there is a positive value of T for
which

x(t ) = x(t + T ) for all t (1.11)

From Eq. (1.11), we can deduce that if x(t ) is periodic with period T, then x(t ) = x(t + mT ) for
all t and for all integers m . Thus, x(t ) is also periodic with period 2T, 3T, …. The fundamental
period T0 of x(t ) is the smallest positive value of T for which Eq. (1.11) holds.
x(t)

...... ......

Fig. 1.8 Continuous-time periodic signal.

5/1 Yao
 ELG 3120 Signals and Systems Chapter 1

A discrete-time signal x[n] is periodic with period N , where N is an integer, if it is unchanged

by a time shift of N,

x[n ] = x[n + N ] (1.12)

for all values of n. If Eq. (1.12) holds, then x[n] is also periodic with period 2 N , 3 N , …. The
fundamental period N0 is the smallest positive value of N for which Eq. (1.12) holds.

x[n]

...... ......
n

Fig. 1.9 Discrete-time periodic signal.

1.2.3 Even and Odd Signals

In addition to their use in representing physical phenomena such as the time shift in a radar
signal and the reversal of an audio tape, transformations of the independent variable are
extremely useful in examining some of the important properties that signal may possess.

Signal with these properties can be even or odd signal, periodic signal:

An important fact is that any signal can be decomposed into a sum of two signals, one of which
is even and one of which is odd.
x(t)
x(t)

t
0

t
0

(a) (b)

Fig. 1.10 An even continuous-time signal; (b) an odd continuous-time signal.

6/1 Yao
 ELG 3120 Signals and Systems Chapter 1

1
EV {x(t)} = [x (t ) + x( −t) ] (1.13)
2

which is referred to as the even part of x(t ) . Similarly, the odd part of x(t ) is given by

1
OD{x (t )} = [x(t ) − x (−t )] (1.14)
2

Exactly analogous definitions hold in the discrete-time case.

1, n ≥ 0 1
x[n] x[n ] =  x[n]
0, n < 0  , n<0
2

EV {x[ n]} =  1, n=0
1
1 1  , n>0
2
1
2
n n

(a) (b)

 1
x[n] − 2 , n < 0

OD{x[n]}=  0, n = 0
 1
 , n>0
 2
1
2

n
1
−
2

(c)

Fig.1.11 The even-odd decomposition of a discrete-time signal.

1.3 Exponential and sinusoidal signals

1.3.1 Continuous-time complex exponential and sinusoidal signals

The continuous-time complex exponential signal

x(t ) = Ce at (1. 15)

where C and a are in general complex numbers.

7/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Real exponential signals

x(t) x(t)

C
C
t t

(a) (b)
Fig. 1.12 The continuous-time complex exponential signal x(t ) = Ce , (a) a > 0 ; (b) a < 0 .
at

Periodic complex exponential and sinusoidal signals

If a is purely imaginary, we have

x(t ) = e j ω0 t (1.16)

An important property of this signal is that it is periodic. We know x(t ) is periodic with period
T if

e j ω0t = e jω0 ( t +T ) = e jω0 t e jω0T (1.17)

For periodicity, we must have

e j ω0T = 1 (1.18)

For ω 0 ≠ 0 , the fundamental period T0 is

2π
T0 = (1.19)
ω0

Thus, the signals e jω0 t and e − jω0 t have the same fundamental period.

A signal closely related to the periodic complex exponential is the sinusoidal signal

x(t ) = A cos(ω 0t + φ) (1.20)

With seconds as the unit of t, the units of φ and ω 0 are radians and radians per second. It is also
known ω 0 = 2πf 0 , where f 0 has the unit of circles per second or Hz.

8/1 Yao
 ELG 3120 Signals and Systems Chapter 1

The sinusoidal signal is also a periodic signal with a fundamental period of T0 .

x ( t ) = A cos(ω 0 t + φ )

2π
T0 =
ω0

A cos φ

Fig. 1.13 Continuous-time sinusoidal signal.

Using Euler’s relation, a complex exponential can be expressed in terms of sinusoidal signals
with the same fundamental period:

e j ω0 t = cosω 0 t + j sin ω 0 t (1.21)

Similarly, a sinusoidal signal can also be expressed in terms of periodic complex exponentials
with the same fundamental period:

A jφ j ω0 t A − jφ − jω 0 t
A cos(ω 0 t + φ ) = e e + e e (1.22)
2 2

A sinusoid can also be expresses as

{
A cos(ω 0 t + φ ) = A Re e j (ω 0t +φ ) } (1.23)

and

{
A sin(ω 0 t + φ ) = A Im e j (ω 0 t +φ ) } (1.24)

Periodic signals, such as the sinusoidal signals provide important examples of signal with infinite
total energy, but finite average power. For example:

E period = ∫ e j ω0 t dt = ∫
T0 T0
1dt = T0 (1.25)
0 0

1
∫ e jω 0t dt = ∫ 1dt = 1
T0 T0
Pperiod = (1.26)
T0 0 0

9/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Since there are an infinite number of periods as t ranges from − ∞ to + ∞ , the total energy
integrated over all time is infinite. The average power is finite since

1 2

∫
T
P∞ = lim e jω 0t dt = 1 (1.27)
T → ∞ 2T −T

Harmonically related complex exponentials:

φ k (t ) = e jk ω0 t , k = 0, ± 1, ± 2, ...... (1.28)

ω 0 is the fundamental frequency.

Example:

Signal x(t ) = e j 2t + e j 3t can be expressed as x(t ) = e j 2.5t (e − j 0.5t + e j 0.5t ) = 2e j 2.5t cos(0.5t ) , the
magnitude of x(t ) is x(t ) = 2 cos(0.5t ) , which is commonly referred to as a full-wave rectified
sinusoid, shown in Fig. 1.14.

x (t )

t
− 4π − 2π 0 2π 4π

Fig. 1.14 Full-wave rectified sinusoid.

General complex Exponential signals

Consider a complex exponential Ce at , where C = C e jθ is expressed in polar and a = r + jω 0 is

expressed in rectangular form. Then

Ce at = C e jθ e ( r + jω 0 )t = C e rt e j (ω 0t +θ ) = C e rt cos(ω 0t + θ ) + j C e rt sin(ω 0t + θ ) . (1.29)

Thus, for r = 0 , the real and imaginary parts of a complex exponential are sinusoidal.
For r > 0 , sinusoidal signals multiplied by a growing exponential.
For r < 0 , sinusoidal signals multiplied by a decaying exponential.

Damped signal – Sinusoidal signals multiplied by decaying exponentials are commonly refereed
to as damped signal.

10/1 Yao
 ELG 3120 Signals and Systems Chapter 1

x(t) x(t)

t t

(a) (b)

Fig. 1.15 (a) Growing sinusoidal signal; (b) decaying sinusoidal signal.

1.3.2 Discrete-time complex exponential and sinusoidal signals

A discrete complex exponential or sequence is defined by

x[n ] = C α n , (1.30)

where C and α are in general complex numbers. This can be alternatively expressed

x[n ] = Ce βn , (1.31)

where α = e β .

Real Exponential Signals

If C and α are real, we have the real exponential signals.

x[n] x[n]

n n

(a) (b)
x[n] x[n]

n n

(c) (d)

11/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Fig. 1.16 Real Exponential Signal x[n ] = C α n : (a) α >1; (b) 0< α <1; (c) –1< α <0; (d) α <-1.

Sinusoidal Signals

x[n ] = e jω0 n (1.32)

e j ω0 n = cos ω 0 n + j sin ω 0 n (1.33)

Similarly, a sinusoidal signal can also be expresses in terms of periodic complex exponentials
with the same fundamental period:

A jφ jω 0n A − jφ − j ω0 n
A cos(ω0 n + φ) = e e + e e (1.34)
2 2

A sinusoid can also be expresses as

{
A cos(ω0 n + φ) = A Re e j ( ω0 n+φ ) } (1.35)

and

{
A sin( ω0 n + φ) = A Im e j (ω0 n +φ ) } (1.36)

The above signals are examples of discrete signals with infinite total energy, but finite average
power. For example: every sample of x[n ] = e jω0 n contributes 1 to the signal’s energy. Thus the
total energy − ∞ < n < +∞ is infinite, while the average power is equal to 1.

12/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Fig.1.17 Discrete-time sinusoidal signal.

General Complex Exponential Signals

Consider a complex exponential Cα n , where C = C e jθ and α = α e jω0 , then

Cα n = C α cos(ω 0 n + θ ) + j C α sin j(ω 0 n + θ ) . (1.37)

n n

Thus, for α = 1 , the real and imaginary parts of a complex exponential are sinusoidal.
For α < 1 , sinusoidal signals multiplied by a decaying exponential.
For α > 1 , sinusoidal signals multiplied by a growing exponential.

13/1 Yao
 ELG 3120 Signals and Systems Chapter 1

(a) (b)

Fig. 1.18 (a) Growing sinusoidal signal; (b) decaying sinusoidal signal.

1.3.3 Periodicity Properties of Discrete-Time Complex Exponentials

There are a number of important differences between continuous-time and discrete-time

sinusoidal signals. The continuous-time signals e jω0 t are distinct for distinct values of ω 0 . For
discrete-time signals, however, these values are not distinct because the signal with ω 0 is
identical to the signals with frequencies ω 0 ± 2π , ω 0 ± 4π , and so on,

e j (ω 0 ±2π ) n = e j (ω 0 ±4π ) n = e j ω0 n . (1.38)

In considering discrete-time exponentials, we need only consider a frequency interval of 2π . In

most occasions, we will use the interval 0 ≤ ω 0 < 2π or − π ≤ ω 0 < π .

The discrete-time signal x[n ] = e jω0 n does not have a continuously increasing rate of oscillation
as ω 0 is increased in magnitude, but as ω 0 is increased from 0, the signal oscillates more and
more rapidly until ω 0 reaches π , and when ω 0 is continuously increased, the rate of oscillation

14/1 Yao
 ELG 3120 Signals and Systems Chapter 1

decreases until ω 0 reaches 2π . We conclude that the low-frequency discrete-time exponentials

have values of ω 0 near 0, 2π , and any other even multiple of π , while the high-frequencies are
located near ω 0 = ±π and other odd multiples of π .

In order for the signal x[n ] = e jω0 n to be periodic with period N > 0 , we must have

e j ω0 (n + N ) = e jω 0 n , (1.39)

or equivalently

e j ω0 N = 1 . (1.40)

For Eq. (1.40) to hold, ω 0 N must be a multiple of 2π . That is, there must be an integer m such
that

ω 0 N = 2πm , (1.41)

or equivalently

ω0 m
= . (1.42)
2π N

From Eq. (1.40), x[n ] = e jω0 n is a periodic if ω 0 / 2π is a rational number and is not periodic
otherwise.

The fundamental frequency of the discrete-time signal x[n ] = e jω0 n is

2π ω 0
= , (1.43)
N m

and the fundamental period of the signal can be

 2π 
N = m .
 (1.44)
ω0 

The comparison of the continuous-time and discrete-time signals are summarized in the table
below:

15/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Table 1 Comparison of the signals e j ω0 t and e j ω 0n .

e j ω0 t e j ω 0n
Distinct signals for distinct values of ω 0 Identical signals for values of ω 0 separated
by multiples of 2π
Periodic for any choice of ω 0 Periodic only if ω 0 = 2πm / N for some
integers N > 0 and m .
Fundamental frequency ω 0 Fundamental frequency ω 0 / m
Fundamental period Fundamental period
ω 0 = 0 : undefined ω 0 = 0 : undefined
2π  2π 
ω0 ≠ 0 : ω 0 ≠ 0 : m  
ω0  ω0 

Example: Suppose that we wish to determine the fundamental period of the discrete-time signal

x[n ] = e j ( 2π / 3) n + e j ( 3π / 4 )n (1.45)

Solution:

The first exponential on the right hand side has a fundamental period of 3. The second
exponential has a fundamental period of 8.

For the entire signal to repeat, each of the terms in Eq. (1.45) must go through an integer number
of its own fundamental period. The smallest increment of n the accomplished this is 24. That is,
over an interval of 24 points, the first term will have gone through 8 of its fundamental periods,
and the second term through three of its fundamental periods, and the overall signal through
exactly one of its fundamental periods.

Harmonically related periodic exponentials

φ k [n ] = e jk ( 2π / N ) n , k = 0, ± 1, ...... (1.46)

In the continuous-time case, all of the harmonically related complex exponentials e jk ( 2π / N )t ,

k = 0, ± 1, ...... , are distinct. But this is not the case for discrete-time signals:

φ k + N [n] = e j ( k +N )( 2π / N )n = e j ( k 2π / N ) n e j 2πn = φ k [n] (1.47)

There are only N distinct period exponentials in the set given in Eq. (1.46).

16/1 Yao
 ELG 3120 Signals and Systems Chapter 1

1.4 The Unit Impulse and Unit Step Functions

The unit impulse and unit step functions in continuous and discrete time are considerably
important in signal and system analysis.

1.4.1 The discrete-Time Unit Impulse and Unit Step Sequences

Discrete-time unit impulse is defined as

0, n≠0
δ [n] =  , (1.48)
1, n=0

δ [n ]

Fig. 1.19 Discrete-time unit impulse.

Discrete-time unit step is defined as

0, n<0
u[n ] =  , (1.49)
1, n≥0

u [ n]
1

n
0

Fig. 1.20 Discrete-time unit step sequence.

The discrete-time impulse unit is the first difference of the discrete-time step

δ [n] = u[n] − u[n − 1] , (1.50)

The discrete-time unit step is the running sum of the unit sample:

17/1 Yao
 ELG 3120 Signals and Systems Chapter 1

n
u[n ] = ∑δ [m] ,
m = −∞
(1.51)

It can be seen that for n < 0 , the running sum is zero, and for n ≥ 0 , the running sum is 1.

∞
If we change the variable of summation from m to k = n − m we have, u[n ] = ∑δ [ n − k ] .
k =0

The unit impulse sequence can be used to sample the value of a signal at n = 0 . Since δ [n] is
nonzero only for n = 0 , it follows that

x[n ]δ [n] = x[0]δ [n] . (1.52)

More generally, a unit impulse δ [n − n0 ] , then

x[n ]δ [n − n 0 ] = x[n0 ]δ [n − n 0 ] (1.53)

This sampling property is very important in signal analysis.

1.4.2 The Continuous-Time Unit Step and Unit Impulse Functions

Continuous-time unit step is defined as

0, t<0
u (t ) =  , (1.54)
1, t≥0

u (t )
1

t
0

Fig. 1.21 Continuous-time unit step function.

The continuous-time unit step is the running integral of the unit impulse

u (t ) = ∫ δ (τ )dτ .
t
(1.55)
−∞

The continuous-time unit impulse can also be considered as the first derivative of the continuous-
time unit step,

18/1 Yao
 ELG 3120 Signals and Systems Chapter 1

du (t )
δ (t ) = . (1.56)
dt

Since u (t ) is discontinuous at t = 0 and consequently is formally not differentiable. This can be

interpreted, however, by considering an approximation to the unit step u ∆ (t ) , as illustrated in the
figure below, which rises from the value of 0 to the value 1 in a short time interval of length ∆ .

u ∆ (t )
δ ∆ (t )

1
1
∆

t t
0 ∆ 0 ∆
(a) (b)

Fig. 1.22 (a) Continuous approximation to the unit step u ∆ (t ) ; (b) Derivative of u ∆ (t ) .

The derivative is

du ∆ (t )
δ ∆ (t ) = , (1.57)
dt

 1
, 0≤t<∆
δ ∆ (t ) =  ∆ , (1.58)
 0, otherwise

as shown in Fig. 1.22.

Note that δ ∆ (t ) is a short pulse, of duration ∆ and with unit area for any value of ∆ . As ∆ → 0 ,
δ ∆ (t ) becomes narrower and higher, maintaining its unit area. At the limit,

δ (t ) = limδ ∆ (t) , (1.59)

∆ →0

u (t ) = lim u ∆ (t ) , (1.60)
∆ →0

and

19/1 Yao
 ELG 3120 Signals and Systems Chapter 1

du (t )
δ (t ) = . (1.61)
dt

Graphically, δ (t) is represented by an arrow pointing to infinity at t = 0 , “1” next to the arrow
represents the area of the impulse.

δ (t ) k δ (t )

1 k

t t
0 0

Fig. 1.23 Continuous-time unit impulse.

Sampling property of the continuous-time unit impulse:

x(t )δ (t) = x (0)δ (t) , (1.62)

Or more generally,

x(t )δ (t − t0 ) = x (t 0 )δ (t − t 0 ) (1.63)

Example:

Consider the discontinuous signal x(t )

x& (t )

2
x (t )
1

2 t
0
1 -1

t -2
0
-1 -3

Fig. 1.24 The discontinuous signal and its derivative.

20/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Note that the derivative of a unit step with a discontinuity of size of k gives rise to an impulse of
area k at the point of discontinuity.

1.5 Continuous-Time and Discrete-Time Systems

A system can be viewed as a process in which input signals are transformed by the system or
cause the system to respond in some way, resulting in other signals as outputs.

Examples

+
v s (t ) +
- C v 0 (t )
i (t )
-

(a)

f (t )

(b)

Fig. 1. 25 Examples of systems. (a) A system with input voltage vs (t ) and output voltage v0 (t ) .
(b) A system with input equal to the force f (t ) and output equal to the velocity v(t ) .

A continuous-time system is a system in which continuous-time input signals are applied and
results in continuous-time output signals.

Continuous-time
x (t ) y (t )
system

A discrete-time system is a system in which discrete-time input signals are applied and results in
discrete-time output signals.

Discrete-time
x[n ] y[n ]
system

21/1 Yao
 ELG 3120 Signals and Systems Chapter 1

1.5.1 Simple Examples of Systems

Example 1: Consider the RC circuit in Fig. 25 (a).

The current i (t ) is proportional to the voltage drop across the resistor:

v ( t ) − vC ( t )
i (t ) = s . (1.64)
R

The current through the capacitor is

dv C (t )
i (t ) = C . (1.65)
dt

Equating the right-hand sides of Eqs. 1.64 and 1.65, we obtain a differential equation describing
the relationship between the input and output:

dvC (t ) 1 1
+ vC ( t ) = v s (t ) , (1.66)
dt RC RC

Example 2: Consider the system in Fig. 25 (b), where the force f (t ) as the input and the velocity
v(t ) as the output. If we let m denote the mass of the car and ρv the resistance due to friction.
Equating the acceleration with the net force divided by mass, we obtain

dv(t ) 1 dv(t ) ρ 1
= [ f (t ) − ρv(t)] ⇒ + v (t ) = f (t ) . (1.67)
dt m dt m m

Eqs.1.66 and 1.77 are two examples of first-order linear differential equations of the form:

dy(t )
+ ay (t ) = bx(t ) . (1.66)
dt

Example 3: Consider a simple model for the balance in a bank account from month to month.
Let y[n] denote the balance at the end of nth month, and suppose that y[n] evolves from month
to month according the equation:

y[ n] = 1.01y[n − 1] + x[n ] , (1.67)

or

y[ n] − 1.01 y[n − 1] = x[n] , (1.68)

where x[n] is the net deposit (deposits minus withdraws) during the nth month 1.01 y[n − 1]
models the fact that we accrue 1% interest each month.

22/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Example 4: Consider a simple digital simulation of the differential equation in Eq. (1.67), in
which we resolve time into discrete intervals of length ∆ and approximate dv(t ) / d (t ) at t = n∆
by the first backward difference, i.e.,

v (n∆ ) − v((n − 1)∆ )

∆

Let v[n] = v (n∆ ) and f [n] = f (n∆ ) , we obtain the following discrete-time model relating the
sampled signals v[n ] and f [n ] ,

m ∆
v[n] − v[n − 1] = f [n] . (1.69)
( m + ρ∆ ) ( m + ρ∆ )

Comparing Eqs. 1.68 and 1.69, we see that they are two examples of the first-order linear
difference equation, that is,

y[ n] + ay[n − 1] = bx[n ] . (1.70)

Some conclusions:

• Mathematical descriptions of systems have great deal in common;

• A particular class of systems is referred to as linear, time-invariant systems.
• Any model used in describing and analyzing a physical system represents an idealization of
the system.

1.5.2 Interconnects of Systems

Input System1 System1 Output

(a)

System1

Input + Output

System 2

(b)

23/1 Yao
 ELG 3120 Signals and Systems Chapter 1

System1 System 2

Input + Output

System 3

(c)

Fig. 1.26 Interconnection of systems. (a) A series or cascade interconnection of two systems; (b)
A parallel interconnection of two systems; (c) Combination of both series and parallel systems.

Input + System1 Output

System 2

Fig. 1.27 Feedback interconnection.

Rs
+
Vi A Vo
-
Vs ±

RL
Vf •
R2
R1

(a)

+ vi = vs − v f BASIC
v L = A vi
vs + AMPLIFIER
A
-

Feedback
Signal FEEDBACK vL
NETWORK
v f = β vL
FB

(b)

24/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Fig. 1.28 A feedback electrical amplifier.

1.6 Basic System Properties

1.6.1 Systems with and without Memory

A system is memoryless if its output for each value of the independent variable as a given time is
dependent only on the input at the same time. For example:

y[ n] = (2 x[ n] − x 2 [n]) 2 , (1.71)

is memoryless.

A resistor is a memoryless system, since the input current and output voltage has the relationship:
i (t )
+
v(t) = R i(t) , (1.72)
v (t )
where R is the resistance. -

One particularly simple memoryless system is the identity system, whose output is identical to its
input, that is

y (t ) = x(t ) , or y[n ] = x[n]

An example of a discrete-time system with memory is an accumulator or summer.

n n −1
y[ n] = ∑ x[k ] = ∑ x[k ] + x[n] = y[n − 1] + x[n] , or
k = −∞ k = −∞
(1.73)

y[ n] − y[n − 1] = x[n] . (1.74)

Another example is a delay

y[n ] = x[n − 1] . (1.75)

A capacitor is an example of a continuous-time system with memory,

i (t )
+
1 t
v(t ) = ∫ i(τ )dτ , (1.76)
C −∞ v (t )
-

25/1 Yao
 ELG 3120 Signals and Systems Chapter 1

where C is the capacitance.

1.6.2 Invertibility and Inverse System

A system is said to be invertible if distinct inputs leads to distinct outputs.

y[n] Inverse
x[n] System w[n]=x[n]
system

y(t)
x(t) y(t)=2x(t) w(t)=0.5y(t) w(t)=x(t)

n
y(t)
x[n] y[n] = ∑ x[k ] w [ n ] = y [ n ] − y[ n − 1] w[ n ] = x [ n ]
k= − ∞

Fig. 1.29 Concept of an inverse system.

Examples of non-invertible systems:

y[ n] = 0 ,

the system produces zero output sequence for any input sequence.

y (t ) = x 2 (t ) ,

in which case, one cannot determine the sign of the input from the knowledge of the output.

Encoder in communication systems is an example of invertible system, that is, the input to the
encoder must be exactly recoverable from the output.

1.6.3 Causality

A system is causal if the output at any time depends only on the values of the input at present
time and in the past. Such a system is often referred to as being nonanticipative, as the system
output does not anticipate future values of the input.

The RC circuit in Fig. 25 (a) is causal, since the capacitor voltage responds only to the present
and past values of the source voltage. The motion of a car is causal, since it does not anticipate
future actions of the driver.

26/1 Yao
 ELG 3120 Signals and Systems Chapter 1

The following expressions describing systems that are not causal:

y[ n] = x[n ] − x[ n + 1] , (1.77)

and

y (t ) = x(t + 1) . (1.78)

All memoryless systems are causal, since the output responds only to the current value of input.

Example: Determine the Causality of the two systems:

(1) y[ n] = x[− n]
(2) y (t ) = x(t ) cos(t + 1)

Solution: System (1) is not causal, since when n < 0 , e.g. n = −4 , we see that y[−4] = x[ 4] , so
that the output at this time depends on a future value of input.

System (2) is causal. The output at any time equals the input at the same time multiplied by a
number that varies with time.

1.6.4 Stability

A stable system is one in which small inputs leads to responses that do not diverge. More
formally, if the input to a stable system is bounded, then the output must be also bounded and
therefore cannot diverge.

Examples of stable systems and unstable systems:

+
v s (t ) +
- C v0 ( t )
i (t )
- f (t )

(a) (b)

The above two systems are stable system.

n
The accumulator y[ n] = ∑ x[k ] is not stable, since the sum grows continuously even if
k = −∞
x[n] is
bounded.

27/1 Yao
 ELG 3120 Signals and Systems Chapter 1

Check the stability of the two systems:

• S1; y (t ) = tx(t ) ;
• S2: y (t ) = e x (t )

• S1 is not stable, since a constant input x(t ) = 1 , yields y (t ) = t , which is not bounded – no
matter what finite constant we pick, y(t ) will exceed the constant for some t.

• S2 is stable. Assume the input is bounded x(t ) < B , or − B < x (t ) < B for all t. We then see
that y(t ) is bounded e − B < y (t) < e B .

1.6.5 Time Invariance

A system is time invariant if a time shift in the input signal results in an identical time shift in
the output signal. Mathematically, if the system output is y(t ) when the input is x(t ) , a time-
invariant system will have an output of y (t − t 0 ) when input is x(t − t 0 ) .

Examples:

• The system y (t ) = sin[ x(t )] is time invariant.

• The system y[n ] = nx[n] is not time invariant. This can be demonstrated by using
counterexample. Consider the input signal x1 [n ] = δ [n] , which yields y1[n] = 0 . However,
the input x 2 [n] = δ [ n − 1] yields the output y 2 [ n] = nδ [ n − 1] = δ [n − 1] . Thus, while x 2 [n] is
the shifted version of x1 [n] , y 2 [n ] is not the shifted version of y1[ n] .

• The system y (t ) = x( 2t ) is not time invariant. To check using counterexample. Consider

x1 (t ) shown in Fig. 1.30 (a), the resulting output y1 (t ) is depicted in Fig. 1.30 (b). If the
input is shifted by 2, that is, consider x 2 (t ) = x1 (t − 2) , as shown in Fig. 1.30 (c), we obtain
the resulting output y 2 (t ) = x2 (2t ) shown in Fig. 1.30 (d). It is clearly seen that
y 2 (t ) ≠ y1 (t − 2) , so the system is not time invariant.

28/1 Yao
 ELG 3120 Signals and Systems Chapter 1

x1 ( t ) y1 (t ) x 2 (t ) = x 1 (t − 2 )

1 1 1

-2 2 -1 1 0 4
(a) (b) (c)
y 2 (t ) y 2 (t − 2 )

1 1

0 2 1 3
(d) (e)

Fig. 1.30 Inputs and outputs of the system y (t ) = x( 2t ) .

1.6.6 Linearity

The system is linear if

• The response to x1 (t ) + x2 (t ) is y1 (t ) + y 2 (t ) - additivity property

• The response to ax1 (t) is ay1 (t ) - scaling or homogeneity property.

The two properties defining a linear system can be combined into a single statement:

• Continuous time: ax1 (t ) + bx2 (t ) → ay1 (t ) + by 2 (t ) ,

• Discrete time: ax1 [n] + bx2 [n] → ay1 [n] + by 2 [n ] .

Here a and b are any complex constants.

Superposition property: If x k [ n], k = 1, 2, 3, ... are a set of inputs with corresponding outputs
y k [n], k = 1, 2, 3, ... , then the response to a linear combination of these inputs given by

x[n ] = ∑ a k x k [n] = a1 x1[ n] + a 2 x 2 [ n] + a3 x 3 [n] + ... , (1.79)

k

is

29/1 Yao
 ELG 3120 Signals and Systems Chapter 1

y[ n] = ∑ a k yk [n ] = a1 y1[n ] + a 2 y 2 [ n] + a3 y 3 [n] + ... , (1.80)

k

which holds for linear systems in both continuous and discrete time.

For a linear system, zero input leads to zero output.

Examples:

• The system y (t ) = tx(t ) is a linear system.

• The system y (t ) = x 2 (t ) is not a liner system.
• The system y[n ] = Re{x[n]}, is additive, but does not satisfy the homogeneity, so it is not a
linear system.
• The system y[ n] = 2 x[n] + 3 is not linear. y[ n] = 3 if x[n ] = 0 , the system violates the “zero-
in/zero-out” property. However, the system can be represented as the sum of the output of a
linear system and another signal equal to the zero-input response of the system. For system
y[ n] = 2 x[n] + 3 , the linear system is

x[n ] → 2 x[n] ,

and the zero-input response is

y 0 [n] = 3

as shown in Fig. 1.31.

y 0 (t )

x (t ) Linear system + y (t )

Fig. 1.31 Structure of an incrementally linear system. y 0 (t ) is the zero-input response of the
system.

The system represented in Fig. 1.31 is called incrementally linear system. The system responds
linearly to the changes in the input.

The overall system output consists of the superposition of the response of a linear system with a
zero-input response.

30/1 Yao
 ELG 3120 Signals and Systems Chapter 2

Chapter 2 Linear Time-Invariant Systems

2.0 Introduction

• Many physical systems can be modeled as linear time-invariant (LTI) systems

• Very general signals can be represented as linear combinations of delayed impulses.
• By the principle of superposition, the response y[n] of a discrete-time LTI system is the sum
of the responses to the individual shifted impulses making up the input signal x[n] .

2.1 Discrete-Time LTI Systems: The Convolution Sum

2.1.1 Representation of Discrete-Time Signals in Terms of Impulses

A discrete-time signal can be decomposed into a sequence of individual impulses.

Example:
x[n]

1
-2 3
n
-4 -3 -1 1 2 4
-1

Fig. 2.1 Decomposition of a discrete-time signal into a weighted sum of shifted impulses.

The signal in Fig. 2.1 can be expressed as a sum of the shifted impulses:

x[n ] = ... + x[−3]δ [ n + 3] + x[−2]δ [n + 2] + x[−1]δ [ n + 1] + x[0]δ [n ] + x[1]δ [ n − 1] + x[2]δ [n − 2] + ...

(2.1)
or in a more compact form

∞
x[n ] = ∑ x[k ]δ [n − k ] .
k = −∞
(2.2)

This corresponds to the representation of an arbitrary sequence as a linear combination of shifted

unit impulse δ [n − k ] , where the weights in the linear combination are x[k ] . Eq. (2.2) is called
the sifting property of the discrete-time unit impulse.

1/2 Yao
 ELG 3120 Signals and Systems Chapter 2

2.1.2 Discrete-Time Unit Impulse Response and the Convolution – Sum Representation of LTI
Systems

Let hk [n ] be the response of the LTI system to the shifted unit impulse δ [n − k ] , then from the
superposition property for a linear system, the response of the linear system to the input x[n] in
Eq. (2.2) is simply the weighted linear combination of these basic responses:

∞
y[ n] = ∑ x[k ]h [n] .
k = −∞
k (2.3)

If the linear system is time invariant, then the responses to time-shifted unit impulses are all
time-shifted versions of the same impulse responses:

hk [n] = h0 [ n − k ] . (2.4)

Therefore the impulse response h[n ] = h0 [n] of an LTI system characterizes the system
completely. This is not the case for a linear time-varying system: one has to specify all the
impulse responses hk [n ] (an infinite number) to characterize the system.

For the LTI system, Eq. (2.3) becomes

∞
y[ n] = ∑ x[k ]h[n − k ] .
k = −∞
(2.5)

This result is referred to as the convolution sum or superposition sum and the operation on the
right-hand side of the equation is known as the convolution of the sequences of x[n] and h[n] .

The convolution operation is usually represented symbolically as

y[n] = x[k ] ∗ h[n] . (2.6)

2.1.3 Calculation of Convolution Sum

• One way to visualize the convolution sum of Eq. (2.5) is to draw the weighted and shifted
impulse responses one above the other and to add them up.

2/2 Yao
 ELG 3120 Signals and Systems Chapter 2

Example: Consider the LTI system with impulse response h[n] and input x[n] , as illustrated in
Fig. 2. 2.
h[n]

1 1 1

n
0 1 2
x[n]
2

0.5

n
0 1
(a)
The output response based on Eq. (2.5) can be expressed

1
y[ n] = ∑ x[ k ] h[ n − k ] = x[ 0]h[ n − 0] + x[1]h[ n − 1] = 0.5h[ n] + 2h[ n − 1] .
k =0

x[0]h[n]=0.5h[n]
0.5

n
0 1 2

2
x[1]h[n-1]=2h[n-1]

n
0 1 2 3

(b)

2.5 2.5

2
y[n]

0.5

n
0 1 2 3

(c)

Fig. 2.2 (a) The impulse response h[n] of an LTI system and an input x[n] to the system; (b) the
responses to the nonzero values of the input; (c) the overall responses.

3/2 Yao
 ELG 3120 Signals and Systems Chapter 2

• Another way to visualize the convolution sum is to draw the signals x[k ] and h[n − k ] as
functions of k (for a fixed n), multiply them to form the signal g[k ] , and then sum all
values of g[k ] .

Example: Calculate the convolution of x[k ] and h[n] shown in Fig. 2.2 (a).

x[k]
0.5

k
0 1

1
h[n-k], n<0

0 1 2

1
h[0-k], n=0

k
-2 -1 0 1 2

1
h[1-k], n=1

k
-1 0 1

1
h[2-k], n=2

k
0 1 2

1
h[3-k], n=3

k
0 1 2 3

h[n-k], n>3 1

k
0

Fig. 2.3 Interpretation of Eq. (2.5) for the signals x[k ] and h[n] .

4/2 Yao
 ELG 3120 Signals and Systems Chapter 2

For n < 0 , y[ n] = 0
∞
For n = 0 , y[0] = ∑ x[k ]h[0 − k ] = 0.5
k = −∞
∞
For n = 1 , y[1] = ∑ x[k ]h[1 − k ] = 0.5 + 2 = 2.5
k = −∞
∞
For n = 1 , y[ 2] = ∑ x[k ]h[2 − k ] = 0.5 + 2 = 2.5
k = −∞
∞
For n = 1 , y[3] = ∑ x[k ]h[ 2 − k ] = 2
k = −∞
For n > 3 , y[ n] = 0

The resulting output values agree with those obtained in the preceding example.

Example: Compute the response of an LTI system described by its impulse response
α n , 0≤n≤6 1, 0≤n≤4
h[n ] =  to the input signal x[n ] =  .
 0, otherwise 0, otherwise

x[n]
1

n
0 1 2 3 4

h[n], α > 1

n
0 1 2 3 4

To do the analysis, it is convenient to consider five separate intervals:

For n < 0 , there is no overlap between the nonzero portions of x[n] and h[n − k ] , and
consequently, y[ n] = 0.

α n − k , 0≤ k ≤n
For 0 ≤ n ≤ 4 , x[k ]h[n − k ] =  ,
 0, otherwise

5/2 Yao
 ELG 3120 Signals and Systems Chapter 2

n n
 1 − α − n−1  1 − α n +1
Thus, in this interval y[ n] = ∑ α n − k =α n ∑ α − k =α n  −1 
=
k =0 k =0  1−α  1−α

α n − k , 0≤ k ≤4
For 4 < n ≤ 6 , x[k ]h[n − k ] = 
 0, otherwise

( )
1 − α −1
5
α n− 4 − α n +1
∑ (α )
4 4
y[ n] = ∑α n −k
=α n −1 k
=α n
= .
k =0 k =0 1 − α −1 1−α

α n− k , ( n − 6) ≤ k ≤ 4
For 6 < n ≤ 10 , x[k ]h[n − k ] = 
 0, otherwise
4
y[n ] = ∑α
k = n− 6
n −k
.

1 − α n −11 α n− 4 − α 7
∑ (α )
10− n 10 − n
Let r = k − n + 6 , y[ n] = ∑α 6− r
=α 6 −1 r
=α 6

1 − α −1
=
1−α
.
r =0 r =0

For n − 6 > 4 , or n > 10 , there is no overlap between the nonzero portions of x[k ] and h[n − k ] ,
and hence, y[ n] = 0 .

The output is illustrated in the figure below.

y[n]

n
0 1 2 3 4 5 6 7 8 9 10

6/2 Yao
 ELG 3120 Signals and Systems Chapter 2

2.2 Continuous-Time LTI systems: the Convolution Integral

The response of a continuous-time LTI system can be computed by convolution of the impulse
response of the system with the input signal, using a convolution integral, rather than a sum.

2.2.1 Representation of Continuous-Time Signals in Terms of Impulses

A continuous-time signal can be viewed as a linear combination of continuous impulses:

∞
x(t ) = ∫ x(τ )δ (t − τ )dτ . (2.7)
−∞

The result is obtained by chopping up the signal x(t ) in sections of width ∆ , and taking sum

x (t )

t
−∆ 0 ∆ 2∆ 3∆

Recall the definition of the unit pulse δ ∆ (t ) ; we can define a signal xˆ(t ) as a linear combination
of delayed pulses of height x( k∆ )

∞
xˆ(t ) = ∑ x(k∆)δ
k = −∞
∆ (t − k∆) ∆ (2.8)

Taking the limit as ∆ → 0 , we obtain the integral of Eq. (2.7), in which when ∆ → 0

(1) The summation approaches to an integral

(2) k∆ → τ and x( k∆ ) → x (τ )
(3) ∆ → dτ
(4) δ ∆(t − k∆) → δ (t − τ )

Eq. (2.7) can also be obtained by using the sampling property of the impulse function. If we
consider t is fixed and τ is time variable, then we have x(τ )δ (t − τ )
= x (τ )δ (−(τ − t )) = x(t )δ (τ − t ) . Hence

7/2 Yao
 ELG 3120 Signals and Systems Chapter 2

∞ ∞ ∞
∫−∞
x(τ )δ (t − τ )dτ = ∫ x(τ )δ (τ − t )dτ = x (t )∫ δ (τ − t )dτ = x(t ) .
−∞ −∞
(2.9)

As in discrete time, this is the sifting property of continuous-time impulse.

2.2.2 Continuous-Time Unit Impulse Response and the Convolution Integral Representation
of an LTI system

The linearity property of an LTI system allows us to calculate the system response to an input
signal xˆ(t ) using Superposition Principle. Let hˆk∆ (t ) be the pulse response of the linear-varying
system to the unit pulses δ ∆ (t − k∆) for − ∞ < k < +∞ . The response of the system to xˆ(t ) is

∞
yˆ (t ) = ∑ x( k∆ ) h
k = −∞
k∆ (t − k∆ )∆ . (2.10)

Note that the response hˆk∆ (t ) tends to the impulse response hτ (t ) as ∆ → 0 . Then at the limit,
we obtain the response of the system to the input signal x(t ) = lim xˆ (t) :
∆ →0

+∞
y (t ) = lim yˆ (t ) =
∆ →0 ∫
−∞
x (τ )hτ (t )dτ . (2.11)

For an LTI system, the impulse responses hτ (t ) are the same as h0 (t ) , except they are shifted by
τ , that is, hτ (t ) = h0 (t − k ) . Then we may define the unit impulse response of the LTI system

h(t ) = h0 (t ) , (2.12)

and an LTI system is completely determined by its impulse response.

So the response to the input signal x(t ) can be written as a convolution integral:

+∞
y (t ) = ∫ x (τ )h(t − τ ) dτ , (2.13)
−∞

or it can be expressed symbolically

y (t ) = x (t ) ∗ h(t ) . (2.14)

8/2 Yao
 ELG 3120 Signals and Systems Chapter 2

2.2.3 Calculation of convolution integral

The output y(t ) is a weighted integral of the input, where the weight on x(τ ) is h(t − τ ) . To
evaluate this integral for a specific value of t ,

• First obtain the signal h(t − τ ) (regarded as a function of τ with t fixed) from h(τ ) by a
reflection about the origin and a shift to the right by t if t >0 or a shift to the left by t is t <0.
• Then multiply together the signals x(τ ) and h(t − τ ) .
• y(t ) is obtained by integrating the resulting product from τ = −∞ to τ = +∞ .

Example: Let x(t ) be the input to an LTI system with unit impulse response h(t ) , where

x(t ) = e − at u (t ) , a > 0 and h(t ) = u (t) .

Step1: The functions h(τ ) , x(τ ) and h(t − τ ) are depicted

h (τ )

τ
0

x (τ )

τ
0

h( t − τ )

1 t<0

τ
t 0

9/2 Yao
 ELG 3120 Signals and Systems Chapter 2

h( t − τ )

1
t> 0

τ
0 t

Step 2: From the figure we can see that for t < 0 , the product of the product x(τ ) and h(t − τ ) is
zero, and consequently, y(t ) is zero. For t > 0

e − at , t0><0τ < t
x(τ )h (t − τ ) = 
 0, otherwise

Step 3: Compute y(t ) by integrating the product for t > 0 :

1 1
y (t ) = ∫ e −aτ dτ = − e− aτ 0 = (1 − e −at ) .
t t

0 a a

The output of y(t ) for all t is

1
y (t ) = (1 − e −at )u (t) , and is shown in figure below.
a

y (t )

1
a

t
0

Example: Compute the convolution of the two signals below:

1, 0<t <T t , 0 < t < 2T

x( t ) =  and h(t ) = 
0, otherwise  0, otherwise

For this example, it is convenient to calculate the convolution in separate intervals. x(τ ) is
sketched and h(t − τ ) is sketched in each of the intervals:

10/2 Yao
 ELG 3120 Signals and Systems Chapter 2

For t < 0 , and t > 3T , x(τ )h (t −τ ) = 0 for all the values of τ , and consequently y(t ) =0.

For other intervals, the product x(τ )h (t − τ ) can be found in the figure on the next page. Thus for
these three intervals, the integration can be calculated with the result shown below:

11/2 Yao
ELG 3120 Signals and Systems Chapter 2

x (τ )

τ
0 T

h (t − τ )

2T
t< 0

τ
t − 2T t 0

h (t − τ ) x(τ ) h ( t − τ )

2T 2T

0<t <T 0<t<T

t
τ
t − 2T 0 t 0 t

h (t − τ ) x(τ ) h ( t − τ )

2T 2T

T < t < 2T t T < t < 2T

τ
t − 2T 0 t 0 T

h (t − τ ) x(τ ) h ( t − τ )

2T 2T
t−T
2T < t < 3T 2 T < t < 3T

τ
0 t − 2T t 0 T

h (t − τ )

2T
t > 3T

τ
0 t − 2T t

12/2 Yao
 ELG 3120 Signals and Systems Chapter 2

 0, t<0 y (t )
 1 2
 t , 0<t <T
 2
 1 2
y (t ) =  Tt − T , T < t < 2T
2
 1 3
− t + Tt + T 2 ,
2
2T < t < 3T
 2 2
t
 0, t > 3T
 0 T 2T 3T

2.3 Properties of Linear Time-Invariant Systems

LTI systems can be characterized completely by their impulse response. The properties can also
be characterized by their impulse response.

2.3.1 The Commutative Property of LTI Systems

A property of convolution in both continuous and discrete time is a Commutative Operation.

That is

∞ ∞
x[n ] ∗ h[n] = h[n ] ∗ x[n] = ∑ x[k ]h[n − k ] = ∑ h[k ]k[n − k ] ,
k = −∞ k = −∞
(2.15)

∞ ∞
x(t ) ∗ h(t) = h(t ) ∗ x(t ) = ∫ x(τ )h(t − τ )dτ = ∫ h (τ ) x(t − τ )dτ . (2.16)
−∞ −∞

x h y h x y

2.3.2 The Distributive Property of LTI Systems

x ∗ (h1 + h2 ) = x ∗ h1 + x ∗ h2 (2.17)

for both discrete-time and continuous-time systems. The property means that summing the
outputs of two systems is equivalent to a system with an impulse response equal to the sum of
the impulse response of the two individual systems, as shown in the figure below.

13/2 Yao
 ELG 3120 Signals and Systems Chapter 2

h1

x + y

h2

x h 1 +h 2 y

The distributive property of convolution can be exploited to break a complicated convolution

into several simpler ones.

For example, an LTI system has an impulse response h[n ] = u[n ] , with an input
n
1
x[n ] =   u[ n] + 2 n u[ −n] . Since the sequence x[n] is nonzero along the entire time axis. Direct
 2
evaluation of such a convolution is somewhat tedious. Instead, we may use the distributive
property to express y[n] as the sum of the results of two simpler convolution problems. That is,
n
1
x1 [n] =   u[n] , x 2 [n] = 2 n u[− n] , using the distributive property we have
 2

y[ n] = (x1 (t) + x 2 (t ) ) ∗ h(t ) = x1 (t ) ∗ h(t ) + x2 (t ) ∗ h(t ) = y1[n ] + y2 [n]

2.3.3 The Associative Property

x ∗ (h1 ∗ h2 ) = (x ∗ h1 ) ∗ h2 . (2.18)

for both discrete-time and continuous-time systems.

x h1 h2 y

x h 1*h 2 y

• For LTI systems, the change of order of the cascaded systems will not affect the response.

14/2 Yao
 ELG 3120 Signals and Systems Chapter 2

• For nonlinear systems, the order of cascaded systems in general cannot be changed. For
example, a two memoryless systems, one being multiplication by 2 and the other squaring the
input, the outputs are different if the order is changed, as shown in the figure below.

w=2x
x 2 w2 y=4x 2

2
w=x 2
x x 2 y=2x 2

2.3.4 LTI system with and without memory

A system is memoryless if its output at any time depends only on the value of its input at the
same time. This is true for a discrete-time system, if h[n ] = 0 for n ≠ 0 . In this case, the impulse
response has the form

h[n ] = Kδ [n ] , (2.19)

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

y[ n] = Kx[ n] . (2.20)

Otherwise the LTI system has memory.

For continuous-time systems, we have the similar results if it is memoryless:

h(t ) = Kδ (t ) , (2.21)

y (t ) = Kx(t ) . (2.22)

Note that if K = 1 in Eqs. (2.19) and (2.21), the systems become identity systems, with output
equal to the input.

2.3.5 Invertibility of LTI systems

We have seen that a system S is invertible if and only if there exists an inverse system S-1 such
that S -1S is an identity system.

x h h1 y=x

15/2 Yao
 ELG 3120 Signals and Systems Chapter 2

Since the overall impulse response in the figure above is h ∗ h1 , h1 must satisfy for it to be the
impulse response of the inverse system, namely h ∗ h1 = δ .

identity
x y=x
system

Applications - channel equalization: for transmission of a signal over a communication channel

such as telephone line, radio link and fiber, the signal at the receiving end is often processed
through a filter whose impulse response is designed to be the inverse of the impulse response of
the communication channel.
Example: Consider a system with a pure time shifted output y (t ) = x(t − t0 ) .

The impulse response of this system is h(t ) = δ (t − t0 ) , since x(t − t 0 ) = x(t ) ∗ δ (t − t 0 ) , that is,
convolution of a signal with a shifted impulse simply shifts the signal

To recover the signal from the output, that is, to invert the system, all that is required is to shift
the output back. So the inverse system should have a impulse response of δ (t + t0 ) , then

δ (t − t0 ) ∗ δ (t + t0 ) = δ (t )

Example: Consider the LTI system with impulse response h[n ] = u[n ] .

The response of this system to an arbitrary input is

+∞
y[ n] = ∑ x[k ]u[n − k ] .
−∞

Considering that u[n − k ] is 0 for n − k < 0 and 1 for n − k ≥ 0 , so we have

n
y[ n] = ∑ x[k ] .
−∞

This is a system that calculates the running sum of all the values of the input up to the present
time, and is called a summer or accumulator. This system is invertible, and its inverse is given as

y[n ] = x[n] − x[n − 1] ,

It is a first difference operation. The impulse response of this inverse system is

h1[n] = δ [ n] − δ [n − 1] ,

16/2 Yao
 ELG 3120 Signals and Systems Chapter 2

We may check that the two systems are really inverses to each other:

h[n ] * h1[ n] = u[n ] * {δ [n] − δ [n − 1]} = u[n ] − u[n − 1] = δ [n]

2.3.6 Causality for LTI systems

A system is causal if its output depends only on the past and present values of the input signal.
Specifically, for a discrete-time LTI system, this requirement is y[n] should not depend on x[k ]
for k > n . Based on the convolution sum equation, all the coefficients h[n − k ] that multiply
values of x[k ] for k > n must be zero, which means that the impulse response of a causal
discrete-time LTI system should satisfy the condition
h[n ] = 0 , for n < 0 (2.23)

A causal system is causal if its impulse response is zero for negative time; this makes sense as
the system should not have a response before impulse is applied.

A similar conclusion can be arrived for continuous-time LTI systems, namely

h(t ) = 0 , for t < 0 (2.24)

Examples: The accumulator h[n ] = u[n ] , and its inverse h[n ] = δ [n] − δ [n − 1] are causal. The
pure time shift with impulse response y (t ) = x(t − t0 ) for t 0 > 0 is causal, but is not causal
for t 0 < 0 .

2.3.7 Stability for LTI Systems

Recall that a system is stable if every bounded input produces a bounded output.

For LTI system, if the input x[n] is bounded in magnitude

x[n] ≤ B , for all n

If this input signal is applied to an LTI system with unit impulse response h[n] , the magnitude of
the output

+∞ +∞ +∞
y[n ] = ∑ h[ k ]x[ n − k ] ≤ ∑ h[k ] x[n − k ] ≤ B ∑ h[k ]
k = −∞ k = −∞ k = −∞
(2.25)

y[n] is bounded in magnitude, and hence is stable if

17/2 Yao
 ELG 3120 Signals and Systems Chapter 2

+∞

∑ h[k ] < ∞ .
k = −∞
(2.26)

So discrete-time LTI system is stable is Eq. (2.26) is satisfied.

The similar analysis applies to continuous-time LTI systems, for which the stability is equivalent
to

+∞
∫−∞
h (τ ) dτ < ∞ . (2.27)

Example: consider a system that is pure time shift in either continuous time or discrete time.

+∞ +∞
In discrete time, ∑ h[k ] = ∑ δ [ n − n
k = −∞ k = −∞
0 = 1,

+∞ +∞
while in continuous time, ∫
−∞
h(τ ) dτ = ∫
−∞
δ ( t − t 0 ) dτ = 1 ,

and we conclude that both of these systems are stable.

+∞ +∞
Example: The accumulator h[n ] = u[n ] is unstable because ∑ h[k ] = ∑ u[n] = ∞ .
k = −∞ k =0

2.3.8 The Unit Step Response of an LTI System

The step response of an LTI system is simply the response of the system to a unit step. It conveys
a lot of information about the system. For a discrete-time system with impulse response h[n] , the
step response is s[n] = u[n ] ∗ h[n] . However, based on the commutative property of convolution,
s[n] = h[n ] ∗ u[n] , and therefore, s[n] can be viewed as the response to input h[n] of a discrete-
time LTI system with unit impulse response. We know that u[n] is the unit impulse response of
the accumulator. Therefore,

n
s[n] = ∑ h[k ] .
k = −∞
(2.28)

From this equation, h[n] can be recovered from s[n] using the relation

h[n ] = s[n ] − s[n − 1] . (2.29)

It can be seen 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.

18/2 Yao
 ELG 3120 Signals and Systems Chapter 2

Similarly, in continuous time, the step response of an LTI system is the running integral of its
impulse response,

∫
t
s (t ) = h (τ )dτ , (2.30)
−∞

and the unit impulse response is the first derivative of the unit step response,

ds(t)
h(t ) = = s' ( t ) . (2.31)
dt

Therefore, in both continuous and discrete time, the unit step response can also be used to
characterize an LTI system.

2.4 Causal LTI Systems Described by Differential and Difference Equations

This is a class of systems for which the input and output are related through

• A linear constant-coefficient differential equation in continuous time, or

• A linear constant-coefficient difference equation in discrete-time.

2.4.1 Linear Constant-Coefficient Differential Equations

In a causal LTI difference system, the discrete-time input and output signals are related
implicitly through a linear constant-coefficient differential equation.

Let us consider a first-order differential equation,

dy(t )
+ 2 y (t ) = x(t ) , (2.32)
dt

where y(t ) denotes the output of the system and x(t ) is the input.

This equation can be explained as the velocity of a car y(t ) subjected to friction force
proportional to its speed, in which x(t ) would be the force applied to the car.

In general, an Nth-order linear constant coefficient differential equation has the form

N
d k y (t) M d k x( t )
∑ a k dt k = ∑ bk dt k , (2.33)
k =0 k =0

19/2 Yao
 ELG 3120 Signals and Systems Chapter 2

The solution of the differential equation can be obtained if we have the N initial conditions (or
auxiliary conditions) on the output variable and its derivatives.

Recall that the solution to the differential equation is the sum of the homogeneous solution of the
N
d k y (t)
differential equation ∑ a k = 0 (a solution with input set to zero) and of a particular
k =0 dt k
solution (a function that satisfy the differential equation).

Forced response of the system = particular solution (usually has the form of the input signal)
Natural response of the system = homogeneous solution (depends on the initial conditions and
forced response).

dy(t )
Example: Solve the system described by + 2 y (t ) = x(t ) . Given the input is x(t ) = Ke 3t u (t) ,
dt
where K is a real number.

As mentioned above, the solution consists of the homogeneous response and the particular
solution:

y (t ) = y h (t ) + y p (t ) , (2.34)

dy(t )
where the particular solution y p (t ) satisfies + 2 y (t ) = x(t ) and homogenous solution y h (t)
dt
satisfies

dy(t )
+ 2 y (t ) = 0 . (2.35)
dt

For the particular solution for t > 0 , y p (t ) is a signal that has the same form as x(t ) for t > 0 ,
that is

y p (t ) = Ye 3t . (2.36)

dy(t )
Substituting x(t ) = Ke 3t u (t) and y p (t ) = Ye 3t into + 2 y (t ) = x(t ) , we get
dt

3Ye 3t + 2Ye 3t = Ke 3t , (2.37)

Canceling the factor e 3t on both sides, we obtain Y = K / 5 , so that

K 3t
y p (t ) = e , t>0 (2.38)
5

20/2 Yao
 ELG 3120 Signals and Systems Chapter 2

To determine the natural response y h (t) of the system, we hypothesize a solution of the form of
an exponential,

y h (t ) = Ae st . (2.39)

Substituting Eq. (3.38) into Eq. (3.35), we get

Ase st + 2 Ase st = 0 , (2.40)

which holds for s = −2 . With this value of s, Ae −2t is a solution to the homogeneous equation
dy(t )
+ 2 y (t ) = 0 for any choice of A.
dt

Combining the natural response and the forced response, we get the solution to the differential
dy(t )
equation + 2 y (t ) = x(t ) :
dt

K 3t
y (t ) = y h (t ) + y p (t ) = Ae −2t + e , t>0 (2.41)
5

Because the initial condition on y(t ) is not specified, so the response is not completely
determined, as the value of A is not known.

For causal LTI systems defined by linear constant coefficient differential equations, the initial
dy (0) dy N −1 (0)
conditions are always y (0) = = ... = = 0 , which is called initial rest.
dt dt N −1

K K
For this example, the initial rest implies that y (0) = 0 , so that y (0) = A + = 0 ⇒ A = − , the
5 5
solution is

K 3t
y (t ) = ( e − e − 2t ) , t > 0 (2.42)
5

For t < 0 , the condition of initial rest and causality of the system implies that y (t ) = 0 , t < 0 ,
since x(t ) = 0 , t < 0 .

2.4.2 Linear Constant-Coefficient Difference Equations

In a causal LTI difference system, the discrete-time input and output signals are related
implicitly through a linear constant-coefficient difference equation.

21/2 Yao
 ELG 3120 Signals and Systems Chapter 2

In general, an Nth-order linear constant coefficient difference equation has the form

N M

∑ a k y[n − k ] = ∑ bk x[ n − k ] ,
k =0 k =0
(2.43)

The solution of the differential equation can be obtained when we have the N initial conditions
(or auxiliary conditions) on the output variable.

The solution to the difference equation is the sum of the homogeneous solution
N

∑a
k =0
k y[n − k ] = 0 (a solution with input set to zero, or natural response) and of a particular
solution (a function that satisfy the difference equation).

y[ n] = y h [ n] + y p [n] , (2.44)

The concept of initial rest of the LTI causal system described by difference equation means that
x[n ] = 0 , n < n0 implies y[ n] = 0 , n < n0 .

Example: consider the difference equation

1
y[ n] − y[n − 1] = x[n] , (2.45)
2

The equation can be rewritten as

1
y[ n] = y[ n − 1] + x[n ] , (2.46)
2

It can be seen from Eq. (2.46) that we need the previous value of the output, y[ n − 1] , to
calculate the current value.

Suppose that we impose the condition of initial rest and consider the input

x[n ] = Kδ [n ] . (2.47)

Since x[n ] = 0 for n ≤ −1 , the condition of initial rest implies that y[ n] = 0 , for n ≤ −1 , so that
we have as an initial condition: y[ −1] = 0 . Starting from this initial condition, we can solve for
successive values of y[n] for n ≥ 0 :

1
y[0] = y[ −1] + x[0] = K ,
2

22/2 Yao
 ELG 3120 Signals and Systems Chapter 2

1 1
y[1] = y[0] + x[1] = K ,
2 2
2
1 1
y[ 2] = y[1] + x[2] =   K ,
2 2

3
1 1
y[3] = y[2] + x[3] =   K ,
2  2
…

n
1 1
y[ n] = y[n − 1] + x[n] =   K .
2 2

Since for an LTI system, the input-output behavior is completely characterized by its impulse
response. Setting K = 1 , , x[n ] = δ [n] we see that the impulse response for the system is

n
 1
h[n ] =   u[n] . (2.48)
 2

Note that the causal system in the above example has an impulse response of infinite duration. In
fact, if N ≥ 1 in Eq. (2.43), the difference equation is recursive, it is usually the case that the LTI
system corresponding to this equation together with the condition of initial rest will have an
impulse response of infinite duration. Such systems are referred to as infinite impulse response
(IIR) systems.

2.4.3 Block Diagram Representations of 1st-order Systems Described by Differential and

Difference Equations

Block diagram interconnection is very simple and nature way to represent the systems described
by linear constant-coefficient difference and differential equations.

For example, the causal system described by the first-order difference equation is
y[ n] + ay[n − 1] = bx[n ] . (2.49)

It can be rewritten as

y[ n] = − ay[n − 1] + bx[n ]

The block diagram representation for this discrete-time system is show:

23/2 Yao
 ELG 3120 Signals and Systems Chapter 2

b
x[n ] + y[n ]

−a
y [ n − 1]

Three elementary operations are required in the block diagram representation: addition,
multiplication by a coefficient, and delay:

x 2 [n ]

adder
multiplication by
a coefficient
a
x1[ n ] + x1 [ n ] + x 2 [ n ] x[n ] ax[n ]

a unit delay

x[n ] D x[ n − 1]

Consider the block diagram representation for continuous-time systems described by a first-order
differential equation:

dy(t )
+ ay (t) = bx(t ) . (2.48)
dt

Eq. (2.48) can be rewritten as

1 dy (t ) b
y (t ) = − + bx(t ) .
a dt a

Similarly, the right-hand side involves three basic operations: addition, multiplication by a
coefficient, and differentiation:

24/2 Yao
 ELG 3120 Signals and Systems Chapter 2

b /a
x (t ) + y (t )

− 1/ a
dy ( t )
dt

x 2 (t )

adder
multiplication by
a coefficient
a
x1 ( t ) + x1 ( t ) + x 2 ( t ) x (t ) ax (t )]

differentiator

dx (t )
x (t ) D
dt

However, the above representation is not frequently used or the representation does not lead to
practical implementation, since differentiators are both difficult to implemented and extremely
sensitive to errors and noise.

An alternative implementation is to used integrators rather than the differentiators. Eq. (2.48) can
be rewritten as

dy(t )
= bx(t ) − ay (t ) , (2.49)
dt

integrating from − ∞ to t , and assuming y (−∞) = 0 , then we obtain

y (t ) = ∫ [bx(τ ) − ay(τ )]dτ .

t
(2.50)
−∞

In this form, the system can be implemented using the adder and coefficient multiplier, together
with an integrator, as shown in the figure below.

25/2 Yao
 ELG 3120 Signals and Systems Chapter 2

integrator

∫
t
x (t )
∫ −∞
x (τ )d τ

b
x (t ) +
∫ y (t )

−a

The integrator can be readily implemented using operational amplifiers, the above
representations lead directly to analog implementations. This is the basis for both early analog
computers and modern analog computation systems.

Eq. (2.50) can also express in the form

y (t ) = y (t 0 ) + ∫ [bx(τ ) − ay (τ )]dτ ,
t
(2.51)
t0

where we consider integrating Eq. (2.50) from a finite point in time t 0 . It makes clear the fact
that the specification of y(t ) requires an initial condition, namely y (t 0 ) .

Any higher-order systems can be developed using the block diagram for the simplest first-order
differential and difference equations.

26/2 Yao
 ELG 3120 Signals and Systems Chapter 3

Chapter 3 Fourier Series Representation of Period Signals

3.0 Introduction

• Signals can be represented using complex exponentials – continuous-time and discrete-time

Fourier series and transform.
• If the input to an LTI system is expressed as a linear combination of periodic complex
exponentials or sinusoids, the output can also be expressed in this form.

3.1 A Historical Perspective

By 1807, Fourier had completed a work that series of harmonically related sinusoids were useful
in representing temperature distribution of a body. He claimed that any periodic signal could be
represented by such series – Fourier Series. He also obtained a representation for aperidic
signals as weighted integrals of sinusoids – Fourier Transform.

Jean Baptiste Joseph Fourier

3.2 The Response of LTI Systems to Complex Exponentials

It is advantageous in the study of LTI systems to represent signals as linear combinations of

basic signals that possess the following two properties:

• The set of basic signals can be used to construct a broad and useful class of signals.

1/3 Yao
 ELG 3120 Signals and Systems Chapter 3

• The response of an LTI system to each signal should be simple enough in structure to provide
us with a convenient representation for the response of the system to any signal constructed
as a linear combination of the basic signal.

Both of these properties are provided by Fourier analysis.

The importance of complex exponentials in the study of LTI system is that the response of an
LTI system to a complex exponential input is the same complex exponential with only a change
in amplitude; that is

Continuous time: e st → H ( s)e st , (3.1)

Discrete-time: z n → H ( z ) z n , (3.2)

where the complex amplitude factor H (s) or H (z) will be in general be a function of the
complex variable s or z.

A signal for which the system output is a (possible complex) constant times the input is referred
to as an eigenfunction of the system, and the amplitude factor is referred to as the system’s
eigenvalue. Complex exponentials are eigenfunctions.

For an input x(t ) applied to an LTI system with impulse response of h(t ) , the output is

+∞ +∞
y (t ) = ∫ h(τ ) x(t − τ )dτ = ∫ h(τ )e s( t −τ ) dτ
−∞ −∞

, (3.3)
+∞ +∞
= ∫ h (τ )e s (t −τ ) dτ = e st ∫ h(τ )e − sτ dτ
−∞ −∞

+∞
where we assume that the integral ∫−∞
h(τ )e − sτ dτ converges and is expressed as

+∞
H (s ) = ∫ h (τ )e − sτ dτ , (3.4)
−∞

the response to e st is of the form

y (t ) = H (s )e st , (3.5)

It is shown the complex exponentials are eigenfunctions of LTI systems and H (s) for a
specific value of s is then the eigenvalues associated with the eigenfunctions.

Complex exponential sequences are eigenfunctions of discrete-time LTI systems. That is,
suppose that an LTI system with impulse response h[n] has as its input sequence

2/3 Yao
 ELG 3120 Signals and Systems Chapter 3

x[n ] = z n , (3.6)

where z is a complex number. Then the output of the system can be determined from the
convolution sum as

∞ ∞ ∞
y[ n] = ∑ h[k ]x[n − k ] = ∑ h[k ]z
k = −∞ k = −∞
n −k
=z n ∑ h[k ]z
k = −∞
−k
. (3.7)

Assuming that the summation on the right-hand side of Eq. (3.7) converges, the output is the
same complex exponential multiplied by a constant that depends on the value of z . That is,

y[ n] = H ( z ) z n , (3.8)

∞
where H ( z ) = ∑ h[k ]z
k = −∞
−k
. (3.9)

It is shown the complex exponentials are eigenfunctions of LTI systems and H (z) for a
specific value of z is then the eigenvalues associated with the eigenfunctions z n .

The example here shows the usefulness of decomposing general signals in terms of
eigenfunctions for LTI system analysis:

Let x(t ) = a1 e s1t + a 2 e s2 t + a3 e s3 t , (3.10)

from the eigenfunction property, the response to each separately is

a1e s1t → a1H1 (s1 )e s1t

a2e s2t → a2 H 2 ( s2 )e s2t

a3e s3t → a3 H 3 (s3 )e s3t

and from the superposition property the response to the sum is the sum of the responses,

y (t ) = a1H1(s1)es1t + a2 H 2 ( s2 )e s2t + a3H 3( s3 )e s3t , (3.11)

Generally, if the input is a linear combination of complex exponentials,

x( t ) = ∑ a k e s k t , (3.12)
k

the output will be

3/3 Yao
 ELG 3120 Signals and Systems Chapter 3

y (t ) = ∑ ak H ( sk )e skt , (3.13)
k

Similarly for discrete-time LTI systems, if the input is

x[n ] = ∑ a k zkn , (3.14)

k

the output is

y[ n] = ∑ ak H ( z k ) zkn , (3.15)
k

3.3 Fourier Series representation of Continuous-Time Periodic Signals

3.31 Linear Combinations of harmonically Related Complex Exponentials

A periodic signal with period of T ,

x(t ) = x(t + T ) for all t , (3.16)

We introduced two basic periodic signals in Chapter 1, the sinusoidal signal

x(t ) = cosω 0 t , (3.17)

and the periodic complex exponential

j ω 0t
x( t ) = e , (3.18)

Both these signals are periodic with fundamental frequency ω 0 and fundamental period
T = 2π / ω 0 . Associated with the signal in Eq. (3.18) is the set of harmonically related complex
exponentials

φ k (t ) = e jk ω0 t = e jk ( 2π / T )t , k = 0, ± 1, ± 2, ...... (3.19)

Each of these signals is periodic with period of T (although for k ≥ 2 , the fundamental period of
φ k (t ) is a fraction of T ). Thus, a linear combination of harmonically related complex
exponentials of the form

4/3 Yao
 ELG 3120 Signals and Systems Chapter 3

+∞ +∞
x( t ) = ∑a e
k = −∞
k
jkω 0 t
= ∑a e
k = −∞
k
jk ( 2π / T ) t
, (3.20)

is also periodic with period of T .

• k = 0 , x(t ) is a constant.
• k = +1 and k = −1 , both have fundamental frequency equal to ω 0 and are collectively
referred to as the fundamental components or the first harmonic components.
• k = +2 and k = −2 , the components are referred to as the second harmonic components.
• k = + N and k = − N , the components are referred to as the Nth harmonic components.

Eq. (3.20) can also be expressed as

+∞
x( t ) = x * ( t ) = ∑a*
k = −∞
k e − jkω 0t , (3.21)

where we assume that x(t ) is real, that is, x(t ) = x * (t) .

Replacing k by − k in the summation, we have

+∞
x( t ) = ∑a*
k = −∞
−k e jkω 0t , (3.22)

which , by comparison with Eq. (3.20), requires that a k = a * −k , or equivalently

a * k = a −k . (3.23)

To derive the alternative forms of the Fourier series, we rewrite the summation in Eq. (2.20) as

[ ]
+∞
x(t ) = a 0 + ∑ a k e jk ω0 t + a −k e − jk ( 2π / T ) t . (3.24)
k =1

Substituting a * k for a −k , we have

[ ]
+∞
x(t ) = a 0 + ∑ a k e jk ω0 t + a *k e − jk (2π / T )t . (3.25)
k =1

Since the two terms inside the summation are complex conjugate of each other, this can be
expressed as

{ }
+∞
x(t ) = a 0 + ∑ 2 Re a k e jkω 0t . (3.26)
k =1

5/3 Yao
 ELG 3120 Signals and Systems Chapter 3

If a k is expressed in polar from as

a k = Ak e jθ k ,

then Eq. (3.26) becomes

{ }
+∞
x(t ) = a 0 + ∑ 2 Re Ak e j ( kω 0t +θ k ) .
k =1

That is

+∞
x(t ) = a 0 + 2 ∑ Ak cos(kω 0 t + θ k ) . (3.27)
k =1

It is one commonly encountered form for the Fourier series of real periodic signals in continuous
time.

Another form is obtained by writing a k in rectangular form as

a k = Bk + jCk

then Eq. (3.26) becomes

+∞
x(t ) = a 0 + 2 ∑ [B k cos kω 0 t − Ck sin kω 0 t ]. (3.28)
k =1

For real periodic functions, the Fourier series in terms of complex exponential has the following
three equivalent forms:

+∞ +∞
x( t ) = ∑a e
k = −∞
k
jkω 0t
= ∑a e
k = −∞
k
jk ( 2π / T ) t

+∞
x(t ) = a0 + 2∑ Ak cos(kω 0t + θ k )
k =1
+∞
x(t ) = a 0 + 2 ∑ [B k cos kω 0 t − C k sin kω 0 t ]
k =1

6/3 Yao
 ELG 3120 Signals and Systems Chapter 3

3.3.2 Determination of the Fourier Series Representation of a Continuous-Time Periodic

Signal

+∞
Multiply both side of x(t ) = ∑a e
k = −∞
k
jkω 0t
by e − jnω0 t , we obtain

+∞
x(t )e − jn ω0 t = ∑a e
k = −∞
k
jkω 0 t
e − jnω0 t , (3.29)

Integrating both sides from 0 to T = 2π / ω 0 , we have

+∞ +∞

∑ a k  ∫ e jk ω 0t e − jnω 0t dt  = ∑ a k  ∫ e j ( k −n )ω 0t dt  ,
T T T
∫0
x(t )e − jnω 0t dt =
k = −∞ 0  k =−∞  0 
(3.30)

Note that

T , k =n
∫
T
e j ( k −n )ω0 t dt = 
0
0, k≠n

So Eq. (3.30) becomes

1 T
an =
T ∫0
x(t )e − jnω 0 t dt , (3.31)

The Fourier series of a periodic continuous-time signal

+∞ +∞
x (t ) = ∑ ak e
k = −∞
jk ω 0 t
= ∑ k
a e
k = −∞
jk ( 2π / T ) t
(3.32)

1 1
ak =
T ∫ T
x(t )e − jkω 0 t dt =
T ∫T
x (t )e − jk ( 2π / T )t dt (3.33)

Eq. (3.32) is referred to as the Synthesis equation, and Eq. (3.33) is referred to as analysis
equation. The set of coefficient {a k } are often called the Fourier series coefficients of the
spectral coefficients of x(t ) .

The coefficient a 0 is the dc or constant component and is given with k = 0 , that is

7/3 Yao
 ELG 3120 Signals and Systems Chapter 3

1
T ∫T
a0 = x(t ) dt , (3.34)

Example: consider the signal x(t ) = sin ω 0 t .

1 jω 0t 1 − jω 0t
sin ω 0 t = e − e .
2j 2j

Comparing the right-hand sides of this equation and Eq. (3.32), we have

1 1
a1 = , a −1 = −
2j 2j

ak = 0 , k ≠ +1 or − 1

Example: The periodic square wave, sketched in the figure below and define over one period is

1, t < T1
x( t ) =  , (3.35)
0, T1 < t < T / 2

The signal has a fundamental period T and fundamental frequency ω 0 = 2π / T .

x (t )

− 2T −T T − T1 T1 T T 2T
−
2 2

To determine the Fourier series coefficients for x(t ) , we use Eq. (3.33). Because of the
symmetry of x(t ) about t = 0 , we choose − T / 2 ≤ t ≤ T / 2 as the interval over which the
integration is performed, although any other interval of length T is valid the thus lead to the same
result.

For k = 0 ,

1 1 2T1
∫ ∫
T1 T1
a0 = x(t) dt = dt = , (3.36)
T − T1 T − T1 T

For k ≠ 0 , we obtain

8/3 Yao
 ELG 3120 Signals and Systems Chapter 3

T1
1 1
∫
T1
− jkω 0t
ak = e dt =− e − jkω 0 t
T −T1 jkω 0T −T1

2  e jk ω 0T1 − e − jk ω 0T1 
=   (3.37)
kω 0T  2j 

2 sin( kω 0 T1 ) sin( kω 0T1 )

= =
kω 0 T kπ

The above figure is a bar graph of the Fourier series coefficients for a fixed T 1 and several
values of T . For this example, the coefficients are real, so they can be depicted with a single
graph. For complex coefficients, two graphs corresponding to the real and imaginary parts or
amplitude and phase of each coefficient, would be required.

3.4 Convergence of the Fourier Series

If a periodic signal x(t ) is approximated by a linear combination of finite number of

harmonically related complex exponentials

N
x N (t ) = ∑a e
k= − N
k
jk ω 0t
. (3.38)

9/3 Yao
 ELG 3120 Signals and Systems Chapter 3

Let eN (t ) denote the approximation error,

N
e N (t ) = x (t ) − x N (t ) = x (t ) − ∑a e
k =− N
k
jkω 0t
. (3.39)

The criterion used to measure quantitatively the approximation error is the energy in the error
over one period:

∫
2
EN = e N (t ) dt . (3.40)
T

It is shown (problem 3.66) that the particular choice for the coefficients that minimize the energy
in the error is

1
ak = ∫
T T
x(t ) e − jkω 0t dt . (3.41)

It can be seen that Eq. (3.41) is identical to the expression used to determine the Fourier series
coefficients. Thus, if x(t ) has a Fourier series representation, the best approximation using only
a finite number of harmonically related complex exponentials is obtained by truncating the
Fourier series to the desired number of terms.

The limit of E N as N → ∞ is zero.

One class of periodic signals that are representable through Fourier series is those signals which
have finite energy over a period,

∫ x (t ) dt < ∞ ,
2
(3.42)
T

When this condition is satisfied, we can guarantee that the coefficients obtained from Eq. (3.33)
are finite. We define

∞
e(t ) = x (t ) − ∑a e
k = −∞
k
jk ω0 t
, (3.43)

then

∫
2
e(t) dt = 0 , (3.44)
T

10/3 Yao
 ELG 3120 Signals and Systems Chapter 3

The convergence guaranteed when x(t ) has finite energy over a period is very useful. In this
case, we may say that x(t ) and its Fourier series representation are indistinguishable.

Alternative set of conditions developed by Dirichlet that guarantees the equivalence of the signal
and its Fourier series representation:

Condition 1: Over any period, x(t ) must be absolutely integrable, that is

∫T
x (t ) dt < ∞ , (3.45)

This guarantees each coefficient a k will be finite, since

1 1
ak =
T ∫T
x(t)e − jkω 0 t dt = ∫ x(t ) dt < ∞ .
T T
(3.46)

A periodic function that violates the first Dirichlet condition is

1
x( t ) = , 0 < t < 1.
t

Condition 2: In any finite interval of time, x(t ) is of bounded variation; that is, there are no
more than a finite number of maxima and minima during a single period of the signal.

An example of a function that meets Condition1 but not Condition 2:

 2π 
x(t ) = sin   , 0 < t ≤ 1, (3.47)
 t 

Condition 3: In any finite interval of time, there are only a finite number of discontinuities.
Furthermore, each of these discontinuities is finite.

An example that violates this condition is a function defined as

x(t ) = 1 , 0 ≤ t < 4 , x(t ) = 1 / 2 , 4 ≤ t < 6 , x(t ) = 1 / 4 , 6 ≤ t < 7 , x(t ) = 1 / 8 , 7 ≤ t < 7.5 , etc.

The above three examples are shown in the figure below.

11/3 Yao
 ELG 3120 Signals and Systems Chapter 3

The above are generally pathological in nature and consequently do not typically arise in
practical contexts.

Summary:

• For a periodic signal that has no discontinuities, the Fourier series representation converges
and equals to the original signal at all the values of t .
• For a periodic signal with a finite number of discontinuities in each period, the Fourier series
representation equals to the original signal at all the values of t except the isolated points of
discontinuity.

Gibbs Phenomenon:

Near a point, where x(t ) has a jump discontinuity, the partial sums x N (t ) of a Fourier series
exhibit a substantial overshoot near these endpoints, and an increase in N will not diminish the
amplitude of the overshoot, although with increasing N the overshoot occurs over smaller and
smaller intervals. This phenomenon is called Gibbs phenomenon.

12/3 Yao
 ELG 3120 Signals and Systems Chapter 3

A large enough value of N should be chosen so as to guarantee that the total energy in these
ripples is insignificant.

3.5 Properties of the Continuous-Time Fourier Series

Notation: suppose x(t ) is a periodic signal with period T and fundamental frequency ω 0 . Then if
the Fourier series coefficients of x(t ) are denoted by a k , we use the notation

x(t ) ←→
FS
ak ,

to signify the pairing of a periodic signal with its Fourier series coefficients.

13/3 Yao
 ELG 3120 Signals and Systems Chapter 3

3.5.1 Linearity

Let x(t ) and y(t ) denote two periodic signals with period T and which have Fourier series
coefficients denoted by a k and b k , that is

x(t ) ←→
FS
a k and y (t ) ←→
FS
bk ,

then we have

z (t ) = Ax (t ) + By (t ) ←→
FS
c k = Aa k + Bb k . (3.48)

3.5.2 Time Shifting

When a time shift to a periodic signal x(t ) , the period T of the signal is preserved.

If x(t ) ←→
FS
a k , then we have

x(t − t0 ) ←→
FS
e − jk ω0 t ak . (3.49)

The magnitudes of its Fourier series coefficients remain unchanged.

3.4.3 Time Reversal

If x(t ) ←→
FS
a k , then

x( −t ) ←→
FS
a −k . (3.50)

Time reversal applied to a continuous-time signal results in a time reversal of the corresponding
sequence of Fourier series coefficients.

If x(t ) is even, that is x(t ) = x(−t ) , the Fourier series coefficients are also even, a −k = a k .
Similarly, if x(t ) is odd, that is x( −t) = − x (t ) , the Fourier series coefficients are also odd,
a −k = − ak .

3.5.4 Time Scaling

+∞
If x(t ) has the Fourier series representation x(t ) = ∑a e
k = −∞
k
jkω 0t
, then the Fourier series

representation of the time-scaled signal x(αt ) is

14/3 Yao
 ELG 3120 Signals and Systems Chapter 3

+∞
x(αt ) = ∑a e
k = −∞
k
jk (αω0 ) t
. (3.51)

The Fourier series coefficients have not changes, the Fourier series representation has changed
because of the change in the fundamental frequency.

3.5.5 Multiplication

Suppose x(t ) and y(t ) are two periodic signals with period T and that

x(t ) ←→
FS
ak ,

y (t ) ←→
FS
bk .

Since the product x(t ) y(t ) is also periodic with period T, its Fourier series coefficients hk is

∞
x(t ) y (t ) ←→
FS
hk = ∑a b
l = −∞
l k −l . (3.52)

The sum on the right-hand side of Eq. (3.52) may be interpreted as the discrete-time convolution
of the sequence representing the Fourier coefficients of x(t ) and the sequence representing the
Fourier coefficients of y(t ) .

3.5.6 Conjugate and Conjugate Symmetry

Taking the complex conjugate of a periodic signal x(t ) has the effect of complex conjugation
and time reversal on the corresponding Fourier series coefficients. That is, if

x(t ) ←→
FS
ak ,

then

x * (t ) ←→
FS
a * −k . (3.53)

If x(t ) is real, that is, x(t ) = x * (t) , the Fourier series coefficients will be conjugate symmetric,
that is

a −k = a * k . (3.54)

15/3 Yao
 ELG 3120 Signals and Systems Chapter 3

From this expression, we may get various symmetry properties for the magnitude, phase, real
parts and imaginary parts of the Fourier series coefficients of real signals. For example:

• From Eq. (3.54), we see that if x(t ) is real, a 0 is real and a − k = a k .

• If x(t ) is real and even, we have a k = a − k , from Eq. (3.54) a −k = a * k , so a k = a * k ⇒ the
Fourier series coefficients are real and even.
• If x(t ) is real and odd, the Fourier series coefficients are real and odd.

3.5.7 Parseval’s Relation for Continuous-Time periodic Signals

Parseval’s Relation for Continuous-Time periodic Signals is

∞
1
∑ ak
2

T ∫T
2
x ( t ) dt = , (3.55)
k = −∞

Since

1 1
∫ ∫
2 2
jkω 0t
= dt = a k ,
2
a k e dt ak
T T T T

2
so that a k is the average power in the kth harmonic component.

Thus, Parseval’s Relation states that the total average power in a periodic signal equals the sum
of the average powers in all of its harmonic components.

16/3 Yao
 ELG 3120 Signals and Systems Chapter 3

3.5.8 Summary of Properties of the Continuous-Time Fourier Series

Property Periodic Signal Fourier Series

Coefficients

x(t)  Periodic with period T and ak


y (t ) fundamenta l frequency ω0 = 2π / T bk
Linearity Ax (t ) + By (t ) Aa k + Bbk
Time Shifting x( t − t 0 ) e − jkω 0t a k
Frequency shifting e jMω0 t x(t) ak− M
Conjugation x * (t ) a *−k
Time Reversal x(−t ) a−k
Time Scaling x(αt ) , α > 0 (Periodic with period T / α ) ak
Periodic Convolution
∫ x (τ ) y(t − τ )dτ
T
Tak bk
Multiplication ∞
x( t ) y( t )
∑ab
l = −∞
l k −l

Differentiation dx(t ) 2π
jkω 0 ak = jk ak
dt T
Integration  1   
∫
t
x (t )dt (finite valued and periodic only 1
  ak =   ak
jk (2π / T ) 
−∞

if a0 = 0 )  jkω 0  

Conjugate Symmetry for x(t ) real  a k = a * −k

Real Signals  Re{a } = Re{a }
 k −k

Im{ak } = − Im{a −k }
 a k = a −k

 ∠a k = −∠a −k
Real and Even Signals x(t ) real and even ak real and even
Real and Odd Signals x(t ) real and odd ak purely imaginary and
odd
Even-Odd Decomposition
of Real Signals  xe (t ) = Ev{x(t )} [x (t ) real ]
 Re{a k }
 xe (t ) = Od {x(t)} [x (t ) real ]
j Im{a k }
Parseval’s Relation for Periodic Signals
∞
1
x(t) dt = ∑ ak
2
∫
2

T T
k = −∞

17/3 Yao
 ELG 3120 Signals and Systems Chapter 3

Example: Consider the signal g (t ) with a fundamental period of 4.

g (t )

1/ 2

t
−2 −1 1 2
− 1/ 2

The Fourier series representation can be obtained directly using the analysis equation (3.33). We
may also use the relation of g (t ) to the symmetric periodic square wave x(t ) discussed on page
8. Referring to that example, T = 4 and T1 = 1 ,

g (t ) = x(t − 1) − 1 / 2 . (3.56)

The time-shift property indicates that if the Fourier series coefficients of x(t ) are denoted by a k
the Fourier series coefficients of x(t − 1) can be expressed as

bk = a k e − jkπ / 2 . (3.57)

The Fourier coefficients of the dc offset in g (t ) , that is the term –1/2 on the right-hand side of
Eq. (3.56) are given by

0, for k ≠ 0

ck =  1 . (3.58)
− 2 , for k = 0

Applying the linearity property, we conclude that the coefficients for g (t ) can be expressed as

a k e − jkπ / 2 , for k ≠ 0

dk =  1 , (3.59)
a 0 − , for k = 0
 2

sin(πk / 2) jk π / 2
replacing a k = e , then we have
kπ

 sin(πk / 2) − jkπ / 2
 e , for k ≠ 0
d k =  πk . (3.60)
0, for k = 0

18/3 Yao
 ELG 3120 Signals and Systems Chapter 3

Example: The triangular wave signal x(t ) with period T = 4 , and fundamental frequency
ω 0 = π / 2 is shown in the figure below.

x (t )

t
−2 2

The derivative of this function is the signal g (t ) in the previous preceding example. Denoting
the Fourier series coefficients of g (t ) by d k , and those of x(t ) by ek , based on the
differentiation property, we have

d k = jk (π / 2)e k . (3.61)

This equation can be expressed in terms of ek except when k = 0 . From Eq. (3.60),

2d k 2 sin(πk / 2) − jkπ / 2
ek = = e . (3.62)
jkπ j(kπ )
2

For k = 0 , e0 can be simply calculated by calculating the area of the signal under one period and
divide by the length of the period, that is

e0 = 1 / 2 . (3.63)

Example: The properties of the Fourier series representation of periodic train of impulse,

∞
x( t ) = ∑ δ (t − kT ) .
k = −∞
(3.64)

We use Eq. (3.33) and select the integration interval to be − T / 2 ≤ t ≤ T / 2 , avoiding the
placement of impulses at the integration limits.

1 T/2 1
ak = ∫
T −T / 2
δ (t )e − jk ( 2π / T )t dt = .
T
(3.65)

All the Fourier series coefficients of this periodic train of impulse are identical, real and even.

19/3 Yao
 ELG 3120 Signals and Systems Chapter 3

The periodic train of impulse has a straightforward relation to square-wave signals such as g (t )
on page 8. The derivative of g (t ) is the signal q (t ) shown in the figure below,

x (t )

− 2T −T T 2T
g (t )

− 2T −T T − T1 T1 T T 2T
−
2 2

q (t )

T1
T − T1 T
−
2 2

which can also interpreted as the difference of two shifted versions of the impulse train x(t ) .
That is,

q (t ) = x(t + T1 ) − x(t − T1 ) . (3.66)

Based on the time-shifting and linearity properties, we may express the Fourier coefficients b k of
q (t ) in terms of the Fourier series coefficient of a k ; that is

b k = e jk ω 0T1 a k − e − jkω 0T1 ak =

T
[
1 jkω 0T1
e ]
− e − jkω 0T1 , (3.67)

Finally we use the differentiation property to get

bk = jkω 0 c k , (3.68)

where c k is the Fourier series coefficients of g (t ) . Thus

20/3 Yao
 ELG 3120 Signals and Systems Chapter 3

bk 2 j sin( kω 0T1 ) 2 sin( kω 0T1 )

ck = = = , k ≠ 0, (3.69)
jkω 0 jkω 0T kω 0T

c0 can be solve by inspection from the figure:

2T1
c0 = . (3.70)
T

Example: Suppose we are given the following facts about a signal x(t )

1. x(t ) is a real signal.

2. x(t ) is periodic with period T = 4 , and it has Fourier series coefficients a k .
3. a k = 0 for k > 1 .
4. The signal with Fourier coefficients bk = e − jπk / 2 a −k is odd.
1 1
∫
2
5. x (t ) dt =
4 4 2

Show that the information is sufficient to determine the signal x(t ) to within a sign factor.

• According to Fact 3, x(t ) has at most three nonzero Fourier series coefficients a k : a −1 , a 0
and a1 . Since the fundamental frequency ω 0 = 2π / T = 2π / 4 = π / 2 , it follows that

x(t ) = a 0 + a1e jπt / 2 + a −1 e − jπt / 2 . (3.71)

• Since x(t ) is real (Fact 1), based on the symmetry property a 0 is real and a1 = a *−1 .
Consequently,

( ) { }
x(t ) = a 0 + a1e jπt / 2 + a1 e j πt / 2 * = a0 + 2 Re a1 e jπt / 2 . (3.72)

• Based on the Fact 4 and considering the time-reversal property, we note that a −k corresponds
to x(−t ) . Also the multiplication property indicates that multiplication of kth Fourier series
by e − jkπ / 2 corresponds to the signal being shifted by 1 to the right. We conclude that the
coefficients b k correspond to the signal x( −(t − 1)) = x(−t + 1) , which according to Fact 4
must be odd. Since x(t ) is real, x(−t + 1) must also be real. So based the property, the
Fourier series coefficients must be purely imaginary and odd. Thus, b0 = 0 , b−1 = −b1 .
• Since time reversal and time shift cannot change the average power per period, Fact 5 holds
even if x(t ) is replaced by x(−t + 1) . That is

1 1
∫
2
x(−t + 1) dt = . (3.73)
4 4 2

21/3 Yao
 ELG 3120 Signals and Systems Chapter 3

Using Parseval’s relation,

b1 + b−1 = 1 / 2 .
2 2
(3.74)

Since b−1 = −b1 , we obtain b1 = 1 / 2 . Since b1 is known to be purely imaginary, it must be

either b1 = j / 2 or b1 = − j / 2 .

• Finally we translate the conditions on b0 and b1 into the equivalent statement on a 0 and
a1 . First, since b0 = 0 , Fact 4 implies that a 0 = 0 . With k = 1 , this condition implies that
a1 = e − j π / 2 b−1 = − jb−1 = jb1 . Thus, if we take b1 = j / 2 , a1 = −1 / 2 , from Eq. (3.72),
x(t ) = − cos(πt / 2) . Alternatively, if we take b1 = − j / 2 , the a1 = 1 / 2 , and therefore,
x(t ) = cos(πt / 2) .

3.6 Fourier Series Representation of Discrete-Time Periodic Signals

The Fourier series representation of a discrete-time periodic signal is finite, as opposed to the
infinite series representation required for continuous-time periodic signals

3.6.1 Linear Combination of Harmonically Related Complex Exponentials

A discrete-time signal x[n] is periodic with period N if

x[n ] = x[n + N ] . (3.75)

The fundamental period is the smallest positive N for which Eq. (3.75) holds, and the
fundamental frequency is ω 0 = 2π / N .

The set of all discrete-time complex exponential signals that are periodic with period N is given
by

φ k [n] = e jkω 0 n = e jk ( 2π / N ) n , k = 0, ± 1, ± 2, .... , (3.76)

All of these signals have fundamental frequencies that are multiples of 2π / N and thus are
harmonically related.

There are only N distinct signals in the set given by Eq. (3.76); this is because the discrete-time
complex exponentials which differ in frequency by a multiple of 2π are identical, that is,

φ k [n ] = φ k +rN [n] . (3.77)

22/3 Yao
 ELG 3120 Signals and Systems Chapter 3

The representation of periodic sequences in terms of linear combinations of the sequences φ k [n]
is

x[n ] = ∑ ak φ k [n ] = ∑ ak e jkω 0n =∑ a k e jk ( 2π / N ) n . (3.78)

k k k

Since the sequences φ k [n] are distinct over a range of N successive values of k, the summation in
Eq. (3.78) need include terms over this range. We indicate this by expressing the limits of the
summation as k = N . That is,

x[n] = ∑ a φ [ n] = ∑ a e
k= N
k k
k= N
k
jkω 0n
= ∑ ak e jk ( 2π / N ) n .
k= N
(3.79)

Eq. (3.79) is referred to as the discrete-time Fourier series and the coefficients ak as the Fourier
series coefficients.

6.2 Determination of the Fourier Series Representation of a Periodic Signal

The discrete-time Fourier series pair:

x[n] = ∑ a φ [ n] = ∑ a e
k= N
k k
k= N
k
jkω 0 n
= ∑a e
k= N
k
jk ( 2π / N ) n
, (3.80)

∑ ∑
1 − jkω0 n 1 − jk ( 2π / N ) n . (3.81)
ak = x[ n ]e = x[ n ]e
N n= N N n= N

Eq. (3.80) is called synthesis equation and Eq. (3.81) is called analysis equation.

Example: Consider the signal x[n ] = sinω 0 n , (3.82)

x[n] is periodic only if 2π / ω 0 is an integer, or a ratio of integer. For the case the when 2π / ω 0
is an integer N, that is, when

2π
ω0 = , (3.83)
N

x[n] is periodic with the fundamental period N. Expanding the signal as a sum of two complex
exponentials, we get

23/3 Yao
 ELG 3120 Signals and Systems Chapter 3

1 j ( 2π / N )n 1 − j ( 2π / N ) n
x[n ] = e − e , (3.84)
2j 2j

From Eq. (3.84), we have

1 1
a1 = , a−1 = − , (3.85)
2j 2j

and the remaining coefficients over the interval of summation are zero. As discussed previously,
these coefficients repeat with period N.

The Fourier series coefficients for this example with N = 5 are illustrated in the figure below.

When 2π / ω 0 is a ratio of integer, that is, when

2πM
ω0 = , (3.86)
N
Assuming the M and N do not have any common factors, x[n] has a fundamental period of N.
Again expanding x[n] as a sum of two complex exponentials, we have

1 jM ( 2π / N )n 1 − jM ( 2π / N ) n
x[n ] = e − e , (3.87)
2j 2j

From which we determine by inspection that a M = (1 / 2 j ) , a − M = −(1 / 2 j) , and the remaining

coefficients over one period of length N are zero. The Fourier coefficients for this example with
M = 3 and N = 5 are depicted in the figure below.

24/3 Yao
 ELG 3120 Signals and Systems Chapter 3

Example: Consider the signal

 2π   2π   4π π
x[n ] = 1 + sin  n + 3 cos n + cos n+ .
N   N   N 2

Expanding this signal in terms of complex exponential, we have

3 1 3 1 1  1 
x[n ] = 1 + ( + )e j ( 2π / N ) n + ( − )e − j ( 2π / N ) n +  e jπ / 2 e j 2( 2π / N ) n +  e − jπ / 2 e − j 2 ( 2π / N )n .
2 2j 2 2j 2  2 

Thus the Fourier series coefficients for this signal are

a0 = 1 ,
3 1 3 1
a1 = + = − j,
2 2j 2 2
3 1 3 1
a −1 = − = + j,
2 2j 2 2
1
a2 = j ,
2
1
a −2 = − j .
2

with a k = 0 for other values of k in the interval of summation in the synthesis equation. The real
and imaginary parts of these coefficients for N = 10 , and the magnitude and phase of the
coefficients are depicted in the figure below.

25/3 Yao
 ELG 3120 Signals and Systems Chapter 3

Example: Consider the square wave shown in the figure below.

Because x[n ] = 1 for − N 1 ≤ n ≤ N 1 , we choose the length-N interval of summation to include

the range − N 1 ≤ n ≤ N 1 . The coefficients are given

N1
1
ak =
N
∑e
n= − N 1
− jk ( 2π / N ) n
, (3.88)

Let m = n + N 1 , we observe that Eq. (3.88) becomes

26/3 Yao
 ELG 3120 Signals and Systems Chapter 3

1 2 N1 − jk ( 2π / N )(m − N1 ) 1 jk ( 2π / N ) N1 2 N1 − jk (2π / N ) m
ak = ∑ e = e ∑e , (3.89)
N n= 0 N n =0

1 jk ( 2π / N ) N1  1 − e jk 2π ( 2 N1 +1) / N  1 sin[2πk ( N 1 + 1 / 2) / N ]
ak = e  − jk ( 2π / N )
 = , k ≠ 0, ± N , ± 2 N , .... (3.90)
N  1− e  N sin( π k / N )
and

2 N1 + 1
ak = , k = 0, ± N , ± 2 N , .... (3.91)
N

The coefficients a k for 2 N 1 + 1 = 5 are sketched for N = 10, 20, and 40 in the figure below.

The partial sums for the discrete-time square wave for M = 1, 2, 3, and 4 are depicted in the
figure below, where N = 9 , 2 N 1 + 1 = 5 .

We see for M = 4 , the partial sum exactly equals to x[n] . In contrast to the continuous-time
case, there are no convergence issues and there is no Gibbs phenomenon.

27/3 Yao
 ELG 3120 Signals and Systems Chapter 3

3.7 Properties of Discrete-Time Fourier Series

Property Periodic Signal Fourier Series Coefficients

x[n] Periodic with period N and ak

 Periodic with period N
y[ n] fundamenta l frequency ω 0 = 2π bk

Linearity Ax[n] + By[n] Aa k + Bbk

Time Shifting x[n − n0 ] e − jk ( 2π / N ) t ak
Frequency shifting e jM ( 2π / N )n x[n ] ak− M
Conjugation x * [n] a *−k

28/3 Yao
 ELG 3120 Signals and Systems Chapter 3

Time Reversal x[− n] a−k

Time Scaling x[n / m], if n is a multipleof n 1  viewed as periodic 
x(m)[n] =  a  
0, if n is a multipleof n m k  with period mN 
(Periodic with period mN )
Periodic Convolution ∑ x[r ] y[ n − r ]
r =[ N ]
Nak bk

Multiplication x[n] y[n]

∑a b l k −l
l =< N >
Differentiation x[n ] − x[n − 1] (1 − e − jk ( 2π / N )
)a
k
Integration n
 1 
∑ x[k ] (finite valued and periodic
k = −∞

1− e
− jk ( 2π / N )  k

a

only if a0 = 0 )
Conjugate Symmetry for x[n] real  a k = a * −k
Real Signals  Re{a } = Re{a }
 k −k

Im{ak } = − Im{a −k }
 a k = a −k

 ∠a k = −∠a −k
Real and Even Signals x[n] real and even ak real and even
Real and Odd Signals x[n] real and odd ak purely imaginary and odd
 xe [n] = Ev{x[n ]} [x[n ] real ]
Even-Odd Decomposition
of Real Signals Re{a k }

 xe [n] = Od {x[n]} [x[n] real] j Im{a k }
Parseval’s Relation for Periodic
Signals

∑ x[ n] = ∑ ak
1 2 2

T n =< N > n =< N >

3.7.1 Multiplication

x[n] y[n] ←→

FS
∑a b
l =< N >
l k −l
. (3.92)

Eq. (3.92) is analogous to the convolution, except that the summation variable is now restricted
to in interval of N consecutive samples. This type of operation is referred to as a Periodic
Convolution between the two periodic sequences of Fourier coefficients.

The usual form of the convolution sum, where the summation variable ranges from − ∞ to + ∞ ,
is sometimes referred to as Aperiodic Convolution.

29/3 Yao
 ELG 3120 Signals and Systems Chapter 3

3.7.2 First Difference

x[ n] − x[ n − 1] ←→
FS
(
1 − e − jk ( 2π / N ) a k ) . (3.93)

3.7.3 Parseval’s Relation

1
∑ = ∑ ak
2 2.
x [ n ]
T n=< N > k =< N >
(3.94)

3.7.4 Examples

Example: Consider the signal shown in the figure below.

x[n]
2

n
-5 0 5

x1 [n]
1

n
-5 0 5

x2 [n]
1

n
-5 0 5

30/3 Yao
 ELG 3120 Signals and Systems Chapter 3

The signal x[n] may be viewed as the sum of the square wave x1 [n] with Fourier series
coefficients b k and x 2 [n] with Fourier series coefficients c k .

a k = bk + c k , (3.95)

The Fourier series coefficients for x1 [n] is

 1 sin( 3πk / 5)
 5 sin(πk / 5) , for k ≠ 0, ± 5, ± 10, ....
bk =  . (3.96)
 ,3
for k = 0, ± 5, ± 10, ....
 5

The sequence x 2 [n] has only a dc value, which is captured by its zeroth Fourier series
coefficient:

1 4
c0 = ∑ x [n] = 1 ,
5 n= 0 2
(3.97)

Since the discrete-time Fourier series coefficients are periodic, it follows that ck = 1 whenever k
is an integer multiple of 5.

 1 sin( 3πk / 5)
 5 sin(πk / 5) , for k ≠ 0, ± 5, ± 10, ....
ak =  (3.98)
8 , for k = 0, ± 5, ± 10, ....
 5

Example: Suppose we are given the following facts about a sequence x[n] :

1. x[n] is periodic with period N = 6 .

∑
5
2. n= 0
x[n] = 2 .

∑
7
3. n= 2
(−1) n x[n] = 1 .
4. x[n] has minimum power per period among the set of signals satisfying the preceding three
conditions.

1 5 1
• From Fact 2, we have a 0 = ∑
6 n =0
x[ n] = .
3
1 7 1
• Note that (−1) n = e − j πn = e − j ( 2π / 6)3 n , we see from Fact 3 that a 3 =
6
∑ 2
x[ n]e − j 3( 2π / N ) n = .
6
• From Parseval’s relation, the average power in x[n] is

31/3 Yao
 ELG 3120 Signals and Systems Chapter 3

5
P = ∑ ak .
2

k =0

Since each nonzero coefficient contributes a positive amount to P, and since the values of a 0 and
a3 are specified, the value of P is minimized by choosing a1 = a 2 = a 4 = a5 = 0 . It follows that

1 1
x[n ] = a0 + a 3 e jπn = + (−1) n ,
3 6

which is shown in the figure below.

1/2
x[n]

1/6

n
-5 0 5

3.8 Fourier Series and LTI Systems

We have seen that the response of a continuous-time LTI system with impulse response h(t ) to a
complex exponential signal e st is the same complex exponential multiplied by a complex gain:

y (t ) = H (s )e st ,
where

∞
H (s ) = ∫ h (τ )e − sτ dτ , (3.99)
−∞

In particular, for s = jω , the output is y (t ) = H ( jω)e jωt . The complex functions H (s) and
H ( jω ) ?are called the system function (or transfer function) and the frequency response,
respectively.

By superposition, the output of an LTI system to a periodic signal represented by a Fourier series

+∞ +∞
x( t ) = ∑ a k e jkω 0t =
k = −∞
∑a e
k = −∞
k
jk ( 2π / T ) t
is given by

32/3 Yao
 ELG 3120 Signals and Systems Chapter 3

+∞
y (t ) = ∑a
k = −∞
k H ( jkω 0 )e jkω 0t . (3.99)

That is, the Fourier series coefficients b k of the periodic output y(t ) are given by

bk = ak H ( jkω 0 ) , (3.100)

Similarly, for discrete-time signals and systems, response h[n] to a complex exponential signal
e j ωn is the same complex exponential multiplied by a complex gain:

y[ n] = H ( jkω 0 )e jkω 0 n , (3.101)

where

∞
H (e j ω ) = ∑ h[n]e
n = −∞
− jωn
. (3.102)

+3
1
Example: Suppose that the periodic signal x(t ) = ∑a e
k = −3
k
jk 2π t
with a 0 = 1 , a1 = a−1 =
4
,

1 1
a 2 = a −2 = , and a 3 = a −3 = is the input signal to an LTI system with impulse response
2 3

h(t ) = e −t u (t )

To calculate the Fourier series coefficients of the output y(t ) , we first compute the frequency
response:

∞
∞ 1 1
H ( jω ) = ∫ e e −τ − jωτ
dτ = e −τ e −ωτ = , (3.103)
0 1 + jω 0
1 + jω

The output is

+3
y (t ) = ∑b e
k = −3
k
jk 2πt
, (3.104)

where bk = a k H ( jkω 0 ) = ak H ( jk 2π ) , so that

1 1  1 1 
b0 = 0 , b1 =   , b−1 =   ,
4  1 + j 2π  4  1 − j 2π 

33/3 Yao
 ELG 3120 Signals and Systems Chapter 3

1 1  1 1 
b2 =   , b−2 =   ,
4  1 + j 4π  4  1 − j 4π 

1 1  1 1 
b3 =   , b−3 =   .
4  1 + j6π  4  1 − j6π 

Example: Consider an LTI system with impulse response h[n ] = α n u[n] , − 1 < α < 1 , and with
the input

 2πn 
x[n ] = cos . (3.105)
 N 

Write the signal x[n] in Fourier series form as

1 1
x[n ] = e j ( 2π / N ) n + e − j (2π / N ) n .
2 2

Also the transfer function is

( )
∞ ∞
1
H (e j ω ) = ∑ α n e − jωn = ∑ α e − j ω
n
= . (3.106)
n =0 n =0 1 − α e − jω

The Fourier series for the output

y[ n] =
1
2
( ) 1
( )
H e j 2 π / N e j ( 2 π / N ) n + H e − j 2 π / N e − j ( 2 π / N )n
2
. (3.107)
1 1  j ( 2 π / N )n 1  1  − j ( 2π / N ) n
=  e +  e
2  1 − α e − jω  2  1 − α e − jω 

34/3 Yao
 ELG 3120 Signals and Systems Chapter 3

3.9 Filtering

Filtering – to change the relative amplitude of the frequency components in a signal or eliminate
some frequency components entirely.

Filtering can be conveniently accomplished through the use of LTI systems with an appropriately
chosen frequency response.
LTI systems that change the shape of the spectrum of the input signal are referred to as
frequency-shaping filters.

LTI systems that are designed to pass some frequencies essentially undistorted and significantly
attenuate or eliminate others are referred to as frequency-selective filters.

Example: A first-order low-pass filter with impulse response h(t ) = e −t u (t ) cuts off the high
frequencies in a periodic input signal, while low frequency harmonics are mostly left intact. The
frequency response of this filter

+∞ 1
H ( jω ) = ∫
0
e −τ e − jωτ dτ =
1 + jω
. (3.107)

We can see that as the frequency ω increase, the magnitude of the frequency response of the
filter H ( jω ) decreases. If the periodic input signal is a rectangular wave, then the output signal
will have its Fourier series coefficients b k given by

sin( kω 0T1 )
bk = a k H ( jkω 0 ) = , k ≠0 (3.108)
kπ (1 + jkω 0 )

2T1
b0 = a 0 H ( 0) = . (3.109)
T

The reduced power at high frequencies produced an output signal that is smother than the input
signal.

t
−T − T1 T1 T

35/3 Yao
 ELG 3120 Signals and Systems Chapter 3

3.10 Examples of continuous-Time Filters Described By Differential

Equations

In many applications, frequency-selective filtering is accomplished through the use of LTI

systems described by linear constant-coefficient differential or difference equations. In fact,
many physical systems that can be interpreted as performing filtering operations are
characterized by differential or difference equation.

3.10.1 A simple RC Lowpass Filter

The first-order RC circuit is one of the electrical circuits used to perform continuous-time
filtering. The circuit can perform either Lowpass or highpass filtering depending on what we
take as the output signal.

v r (t )

+
v s (t ) -
v c (t )

If we take the voltage cross the capacitor as the output, then the output voltage is related to the
input through the linear constant-coefficient differential equation:

dvc (t )
RC + vc (t ) = v s (t ) . (3.111)
dt

Assuming initial rest, the system described by Eq. (3.111) is LTI. If the input is vs (t ) = e jωt , we
must have voltage output vc (t ) = H ( jω )e jωt . Substituting these expressions into Eq. (3.111), we
have

RC
d
dt
[ ]
H ( jω )e jωt + H ( jω )e jωt = e j ωt , (3.112)

or

RCj ωH ( jω )e jωt + H ( jω )e jω t = e jωt , (3.113)

36/3 Yao
 ELG 3120 Signals and Systems Chapter 3

1
Then we have H ( jω ) = . (3.114)
1 + RCj ω

Te amplitude and frequency response H ( jω ) is shown in the figure below.

We can also get the impulse response

1 −t / RC
h(t ) = e u (t ) , (3.115)
RC

and the step response is

h(t ) = (1 − e −t / RC )u (t ) , (3.116)

The fundamental trade-off can be found

by comparing the figures:

• To pass only very low frequencies,

1 / RC should be small, or RC should
be large.

• To have fast step response, we need a

smaller RC .

• The type of trade-off between

behaviors in the frequency domain
and time domain is typical of the
issues arising in the design analysis of
LTI systems.

37/3 Yao
 ELG 3120 Signals and Systems Chapter 3

3.10.2 A Simple RC Highpass Filter

If we choose the output from the resistor, then we get an RC highpass filter.

3.11 Examples of Discrete-Time Filter Described by Difference Equations

A discrete-time LTI system described by the first-order difference equation

y[ n] − ay[n − 1] = x[n] . (3.116)

Form the eigenfunction property of complex exponential signals, if x[n ] = e jωn , then
y[ n] = H (e jω )e j ωn , where H (e j ω ) is the frequency response of the system.

1
H (e j ω ) = . (3.117)
1 − ae − jω

The impulse response of the system is

x[n ] = a n u[n] . (3.118)

The step response is

1 − a n+1
s[n] = u[n ] . (3.119)
1− a

From the above plots we can see that for a = 0.6 the system acts as a Lowpass filter and
a = −0.6 , the system is a highpass filter. In fact, for any positive value of a < 1 , the system
approximates a highpass filter, and for any negative value of a > −1 , the system approximates a

38/3 Yao
 ELG 3120 Signals and Systems Chapter 3

highpass filter, where a controls the size of bandpass, with broader pass bands as a in
decreased.

The trade-off between time domain and frequency domain characteristics, as discussed in
continuous time, also exists in the discrete-time systems.

3.11.2.2 Nonrecursive Discrete-Time Filters

The general form of an FIR norecursive difference equation is

M
y[ n] = ∑ b x[n − k ] .
k =− N
k (3.120)

It is a weighted average of the (N + M + 1) values of x[n] , with the weights given by the
coefficients b k .

One frequently used example is a moving-average filter, where the output of y[n] is an average
of values of x[n] in the vicinity of n0 - the result corresponding a smooth operation or lowpass
filtering.

An example: y[ n] =
1
(x[ n − 1] + x[n] + x[ n + 1]) . (3.121)
3

The impulse response is

h[n ] =
1
(δ [n − 1] + δ [ n] + δ [n + 1]) , (3.122)
3

and the frequency response

H (e j ω ) =
3
(
1 jω
e + 1 + e − jω .) (3.123)

39/3 Yao
 ELG 3120 Signals and Systems Chapter 3

A generalized moving average filter can be expressed as

M
1
y[ n] = ∑ b x[ n − k ] .
N + M + 1 k =− N k
(3.124)

The frequency response is

1 M
1 sin[ω (M + N + 1) / 2]
jω
H (e ) = ∑
M + N + 1 k =− N
e − jω k =
M + N +1
e j ω [ ( N − M ) / 2]
sin(ω / 2)
. (3.125)

The frequency responses with different average window lengths are plotted in the figure below.

FIR norecursive highpass filter

An example of FIR norecursive highpass filter is

40/3 Yao
 ELG 3120 Signals and Systems Chapter 3

x[n] − x[n − 1]
y[ n] = . (3.126)
2

The frequency response is

H (e j ω ) =
1
2
( )
1 − e − jω = je jω / 2 sin(ω / 2) . (3.127)

41/3 Yao
 ELG 3120 Signals and Systems Chapter 4

Chapter 4 Continuous-Time Fourier Transform

4.0 Introduction

• A periodic signal can be represented as linear combination of complex exponentials which

are harmonically related.
• An aperiodic signal can be represented as linear combination of complex expone ntials, which
are infinitesimally close in frequency. So the representation take the form of an integral
rather than a sum
• In the Fourier series representation, as the period increases the fundamental frequency
decreases and the harmonically related components become closer in frequency. As the
period becomes infinite, the frequency components form a continuum and the Fourier series
becomes an integral.

4.1 Representation of Aperiodic Signals: The Continuous-Time Fourier

Transform

4.1.1 Development of the Fourier Transform Representation of an Aperiodic Signal

Starting from the Fourier series representation for the continuous-time periodic square wave:

1, t < T1
x( t ) =  , (4.1)
0, T1 < t < T / 2

x (t )

− 2T −T T − T1 T1 T T 2T
−
2 2
The Fourier coefficients a k for this square wave are

2 sin( kω 0 T1 )
ak = . (4.2)
kω 0 T

or alternatively

1/4 Yao
 ELG 3120 Signals and Systems Chapter 4

2 sin(ωT1 )
Ta k = , (4.3)
ω ω = kω0

where 2 sin(ωT1 ) / ω represent the envelope of Ta k

• When T increases or the fundamental frequency ω 0 = 2π / T decreases, the envelope is

sampled with a closer and closer spacing. As T becomes arbitrarily large, the original
periodic square wave approaches a rectangular pulse.

• Ta k becomes more and more closely spaced samples of the envelope, as T → ∞ , the Fourier
series coefficients approaches the envelope function.

This example illustrates the basic idea behind Fourier’s development of a representation for
aperiodic signals.

Based on this idea, we can derive the Fourier transform for aperiodic signals.

Suppose a signal x(t ) with a finite duration, that is, x(t ) = 0 for t > T1 , as illustrated in the
figure below.

• From this aperiodic signal, we construct a periodic signal ~

x (t) , shown in the figure below.

2/4 Yao
 ELG 3120 Signals and Systems Chapter 4

• As T → ∞ , ~
x (t ) = x(t) , for any infinite value of t .

• The Fourier series representation of ~

x (t) is

+∞
~
x (t ) = ∑a e
k = −∞
k
jkω 0t
, (4.4)

1 T/2 ~
ak =
T ∫−T / 2
x (t )e − jk ω0 t dt . (4.5)

• Since ~
x (t ) = x(t) for t < T / 2 , and also, since x(t ) = 0 outside this interval, so we have

1 T/2 1 ∞
ak =
T ∫−T / 2
x(t)e − jkω 0 t dt = ∫ x (t )e − jkω 0t dt .
T −∞

• Define the envelope X ( jω ) of Ta k as

∞
X ( jω ) = ∫ −∞
x(t )e − jωt dt . (4.6)

we have for the coefficients a k ,

1
ak = X ( jkω 0 )
T

Then ~
x (t) can be expressed in terms of X ( jω ) , that is

+∞ +∞
1 1
~
x (t ) = ∑
k = −∞ T
X ( jkω 0 )e jk ω 0t =
2π
∑ X ( jkω
k = −∞
0 )e jk ω0 t ω 0 . (4.7)

3/4 Yao
 ELG 3120 Signals and Systems Chapter 4

• As T → ∞ , ~
x (t ) = x(t) and consequently, Eq. (4.7) becomes a representation of x(t ) .

• In addition, ω 0 → 0 as T → ∞ , and the right-hand side of Eq. (4.7) becomes an integral.

We have the following Fourier transform:

1 ∞
∫−∞
jω t
x( t ) = X ( j ω ) e dω Inverse Fourier Transform
2π (4.8)

and

∞
X ( j ω ) = ∫ x(t )e − jωt dt Fourier Transform (4.9)
−∞

4.1.2 Convergence of Fourier Transform

If the signal x(t ) has finite energy, that is, it is square integrable,

∞ 2

∫−∞
x (t ) dt < ∞ , (4.10)

Then we guaranteed that X ( jω ) is finite or Eq. (4.9) converges. If e(t ) = ~

x (t ) − x (t ) , we have

∞ 2

∫−∞
e(t ) dt = 0 . (4.11)

An alterative set of conditions that are sufficient to ensure the convergence:

Contition1: Over any period, x(t ) must be absolutely integrable, that is

∞
∫−∞
x (t ) dt < ∞ , (4.12)

Condition 2: In any finite interval of time, x(t ) have a finite number of maxima and mi nima.

Condition 3: In any finite interval of time, there are only a finite number of discontinuities.
Furthermore, each of these discontinuities is finite.

4/4 Yao
 ELG 3120 Signals and Systems Chapter 4

4.1.3 Examples of Continuous-Time Fourier Transform

Example: consider signal x(t ) = e −at u (t) , a > 0 .

From Eq. (4.9),

∞
∞ 1 1
X ( jω ) = ∫ e e − at − j ωt
dt = − e −( a+ jω )t = , a >0 (4.12)
0 a + jω 0
a + jω

If a is complex rather then real, we get the same result if Re{a} > 0

The Fourier transform can be plotted in terms of the magnitude and phase, as shown in the figure
below.

1 ω 
X ( jω ) = , ∠X ( jω ) = − tan −1   . (4.13)
a +ω2 2
a

Example: Let x(t ) = e − a t , a >0

∞ ∞ 1 1 2a
∫ ∫ e at e − j ωt dt + ∫ e −at e − jωt dt =
−a t 0
X ( jω ) = e e − jωt dt = + = 2
−∞ −∞ 0 a − jω a + jω a + ω 2

The signal and the Fourier transform are sketched in the figure below.

5/4 Yao
 ELG 3120 Signals and Systems Chapter 4

Example: x(t ) = δ (t ) . (4.14) x( t ) = δ ( t ) X ( jω ) = 1

∞
X ( jω ) = ∫
−∞
δ (t )e − jωt dt = 1 . (4.15)

That is, the impulse has a Fourier transform consisting of equal contributions at all frequencies.

Example: Calculate the Fourier transform of the rectangular pulse signal

1, t < T1
x( t ) =  . (4.16)
0, t > T1
x (t )

− T1 T1

∞ sin ωT1
∫ ∫
T1
X ( jω ) = x(t )e − j ωt dt = 1e − jωt dt = 2 .
−∞ −T1 ω
(4.17)

The Inverse Fourier transform is

1 ∞ sin ωT1 jωt

xˆ(t ) =
2π ∫ −∞
2
ω
e dω , (4.18)

Since the signal x(t ) is square integrable,

∞ 2
e(t ) = ∫ x(t ) − xˆ(t ) dt = 0 . (4.19)
−∞

xˆ(t ) converges to x(t ) everywhere except at the discontinuity, t = ±T1 , where xˆ(t ) converges to
½, which is the average value of x(t ) on both sides of the discontinuity.

In addition, the convergence of xˆ(t ) to x(t ) also exhibits Gibbs phenomenon. Specifically, the
integral over a finite-length interval of frequencies

1 sin ωT1 jωt

∫
W
2 e dω
2π −W ω

6/4 Yao
 ELG 3120 Signals and Systems Chapter 4

As W → ∞ , this signal converges to x(t ) everywhere, except at the discontinuities. More over,
the signal exhibits ripples near the discontinuities. The peak values of these ripples do not
decrease as W increases, although the ripples do become compressed toward the discontinuity,
and the energy in the ripples converges to zero.

Example: Consider the signal whose Fourier transform is

1, ω <W
X ( jω ) =  .
0, ω >W

The Inverse Fourier transform is

1 sin Wt
∫
W
x( t ) = e jωt dω = .
2π −W πt

Comparing the results in the preceding example and this example, we have

→
FT

Square wave Sinc function

←

FT −1

This means a square wave in the time domain, its Fourier transform is a sinc function. However,
if the signal in the time domain is a sinc function, then its Fourier transform is a square wave.
This property is referred to as Duality Property.

We also note that when the width of X ( jω ) increases, its inverse Fourier transform x(t ) will be
compressed. When W → ∞ , X ( jω ) converges to an impulse. The transform pair with several
different values of W is shown in the figure below.

7/4 Yao
 ELG 3120 Signals and Systems Chapter 4

4.2 The Fourier Transform for Periodic Signals

The Fourier series representation of the signal x(t ) is

∞
x( t ) = ∑a e
k = −∞
k
jkω 0t
. (4.20)

It’s Fourier transform is

∞
X ( jω ) = ∑ 2πa δ (ω −kω
k = −∞
k 0 ).
(4.21)

Example: If the Fourier series coefficients for the square wave below are given

x (t )

− 2T −T T − T1 T1 T T 2T
−
2 2

sin kω 0 T1
ak = , (4.22)
πk

The Fourier transform of this signal is

∞
2 sin kω 0T1
X ( jω ) = ∑
k = −∞ k
δ (ω −kω 0 ) . (4.23)

8/4 Yao
 ELG 3120 Signals and Systems Chapter 4

Example: The Fourier transforms for x(t ) = sin ω 0 t and x(t ) = cosω 0 t are shown in the figure
below.

9/4 Yao
 ELG 3120 Signals and Systems Chapter 4

∞
Example: Calculate the Fourier transform for signal x(t ) = ∑ δ (t − kT ) .
k = −∞

The Fourier series of this signal is

1 +T / 2 1
ak = ∫
T −T / 2
δ (t )e − jω 0t = .
T

The Fourier transform is

2π ∞
2πk
X ( jω ) =
T
∑ δ (ω −
k = −∞ T 0
).

The Fourier transform of a periodic impulse train in the time domain with period T is a periodic
impulse train in the frequency domain with period 2π / T , as sketched din the figure below.

4.3 Properties of The Continuous-Time Fourier Transform

4.3.1 Linearity

If x(t ) ←→
F
X ( jω ) and y (t ) ←→
F
Y ( jω )

Then

10/4 Yao
 ELG 3120 Signals and Systems Chapter 4

ax (t ) + by (t ) ←→
F
aX ( jω ) + bY ( jω ) . (4. 20)

4.3.2 Time Shifting

If x(t ) ←→
F
X ( jω )

Then

x (t − t0 ) ←→
F
e − jω t0 X ( jω ) . (4. 20)

Or

F {x (t − t0 )} = e − jω t0 X ( jω ) = X ( jω ) e j [∠X ( jω )−ω t0 ] . (4. 20)

Thus, the effect of a time shift on a signal is to introduce into its transform a phase shift, namely,
− ω 0t .

Example: To evaluate the Fourier transform of the signal x(t ) shown in the figure below.

x (t )

1.5
1
t
1 2 3 4

x 2 (t ) x1 ( t )

1 1

t t
3 3 1 1
− −
2 2 2 2

The signal x(t ) can be expressed as the linear combination

1
x (t ) = x1 (t − 2.5) + x2 (t − 2.5) . (4. 20)
2

x1 (t ) and x 2 (t ) are rectangular pulse signals and their Fourier transforms are

11/4 Yao
 ELG 3120 Signals and Systems Chapter 4

2 sin(ω / 2) 2 sin( 3ω / 2)
X 1 ( jω ) = and X 2 ( jω ) =
ω ω

Using the linearity and time-shifting properties of the Fourier transform yields

 sin(ω / 2) + 2 sin( 3ω / 2) 
X ( j ω ) = e − j 5ω / 2  
 ω 

4.3.3 Conjugation and Conjugate Symmetry

If x(t ) ←→
F
X ( jω )

Then

x * (t ) ←→
F
X * (− jω ) . (4. 20)

∗
Since X * ( jω ) = ∫ x(t )e − jω t dt  =
+∞ +∞

 − ∞  ∫−∞
x * (t )e j ωt dt ,

Replacing ω by − ω , we see that

+∞
X * (− jω ) = ∫ x * (t )e − jωt dt , (4. 20)
−∞

The right-hand side is the Fourier transform of x * (t ) .

If x(t ) is real, from Eq. (4.20) we can get

X (− jω ) = X * ( jω ) . (4. 20)

We can also prove that if x(t ) is both real and even, then X ( jω ) will also be real and even.
Similarly, if x(t ) is both real and odd, then X ( jω ) will also be purely imaginary and odd.

A real function x(t ) can be expressed in terms of the sum of an even function
xe (t ) = Ev{x(t )}and an odd function xo (t ) = Od {x (t )} . That is

x( t ) = xe ( t ) + xo ( t )

Form the Linearity property,

12/4 Yao
 ELG 3120 Signals and Systems Chapter 4

F {x (t )} = F {xe (t )}+ F {xo (t )},

From the preceding discussion, F {x e (t )} is real function and F {x o (t )} is purely imaginary. Thus
we conclude with x(t ) real,

x(t ) ←→
F
X ( jω )

Ev{x(t)}←→
F
Re{X ( jω )}

Od {x (t )}←→
F
j Im{X ( jω )}

Example: Using the symmetry properties of the Fourier transform and the result
1
e −at u (t) ←→
F
to evaluate the Fourier transform of the signal x(t ) = e − a t , where a > 0 .
a + jω

 e − at u (t) + e at u( −t ) 
Since x (t ) = e − a t = e − at u(t ) + e at u ( −t ) = 2   = 2 Ev{e u (t )},
− at

 2 

 1  2a
So X ( jω ) = 2 Re  = 2
 a + jω  a + ω
2

4.3.4 Differentiation and Integration

If x(t ) ←→
F
X ( jω )

Then

dx(t ) F
←→ jωX ( jω ) . (4. 20)
dt

1
∫
t
x (τ )dτ ←→
F
X ( jω ) + πX (0)δ (ω )
−∞ jω . (4. 20)

Example: Consider the Fourier transform of the unit step x(t ) = u (t) .

It is know that

13/4 Yao
 ELG 3120 Signals and Systems Chapter 4

g (t ) = δ (t ) ←→
F
1

Also note that

x(t ) = ∫ g (τ ) dτ
t

−∞

The Fourier transform of this function is

1 1
X ( jω ) = + πG (0)δ (ω ) = + πδ (ω ) .
jω jω

where G (0) = 1 .

Example: Consider the Fourier transform of the function x(t ) shown in the figure below.

x(t) 1
1
1 −1 1
−1 t
t t
1 −1 −1
= 1
+ −1

dx(t )
g (t ) =
dt

From the above figure we can see that g (t ) is the sum of a rectangular pulse and two impulses.

 2 sin ω 
G ( jω ) =   − e jω − e − jω
 ω 

Note that G (0) = 0 , using the integration property, we obtain

G( jω ) 2 sin ω 2 cos ω
X ( jω ) = + πG(0)δ (ω ) = − .
jω jω 2 jω

It can be found X ( jω ) is purely imaginary and odd, which is consistent with the fact that x(t ) is
real and odd.

4.3.5 Time and Frequency Scaling

14/4 Yao
 ELG 3120 Signals and Systems Chapter 4

x(t ) ←→
F
X ( jω ) ,

Then
1 jω
x( at) ←→
F
X( ). (4. 20)
a a

From the equation we see that the signal is compressed in the time domain, the spectrum will be
extended in the frequency domain. Conversely, if the signal is extended, the corresponding
spectrum will be compressed.

If a = −1 , we get from the above equation,

x( −t ) ←→
F
X (− jω ) . (4. 20)

That is, reversing a signal in time also reverses its Fourier transform.

4.3.6 Duality

The duality of the Fourier transform can be demonstrated using the following example.

1, t < T1 2 sin ωT1

x1 (t ) =  ←→ X 1 ( jω ) =
F

0, t > T1 ω

sin WT1 F 1, ω <W

x 2 (t ) = ←→ X 2 ( jω ) = 
πt 0, ω >W

15/4 Yao
 ELG 3120 Signals and Systems Chapter 4

The symmetry exhibited by these two examples extends to Fourier transform in general. For any
transform pair, there is a dual pair with the time and frequency variables interchanged.

2
Example: Consider using duality and the result e − t ←→
F
X ( jω ) = to find the Fourier
1+ ω 2
transform G ( jω ) of the signal

2
g (t ) = .
1+ t 2

−t 2
Since e ←→
F
X ( jω ) = , that is,
1+ ω 2

1 ∞  2  jωt
∫
−t
e =   e dω ,
2π −∞ 1 + ω 2
 

Multiplying this equation by 2π and replacing t by − t , we have

∞  2  − jωt
=∫ 
−t
2πe  e dω
−∞ 1 + ω 2
 
Interchanging the names of the variables t and ω , we find that

∞  2  − jωt  2 
=∫ 
−ω −ω
2πe  e dω ⇒ F −1   = 2πe .
−∞ 1 + t
  1 + t 
2 2

Based on the duality property we can get some other properties of Fourier transform:

dX ( jω )
− jtx (t ) ←→
F

dω

e jω 0t x(t ) ←→
F
X ( j (ω − ω 0 ))

1 ω
− x (t ) + πx( 0)δ (t ) ←→ ∫ x(η ) dη
F

jt − ∞

16/4 Yao
 ELG 3120 Signals and Systems Chapter 4

4.3.7 Parseval’s Relation

If x(t ) ←→
F
X ( jω ) ,

We have

∞ 1 ∞
∫ x (t ) dt = ∫ X ( jω ) dω
2 2
−∞ 2π −∞

Parseval’s relation states that the total energy may be determined either by computing the energy
2
per unit time x(t ) and integrating over all time or by computing the energy per unit frequency
X ( jω ) / 2π and integrating over all frequencies. For this reason, X ( jω ) is often referred to
2 2

as the energy-density spectrum.

4.4 The convolution properties

y (t ) = h(t ) ∗ x (t ) ←→
F
Y ( jω ) = H ( jω ) X ( jω )
The equation shows that the Fourier transform maps the convolution of two signals into product
of their Fourier transforms.

H ( jω ) , the transform of the impulse response, is the frequency response of the LTI system,
which also completely characterizes an LTI system.

Example: The frequency response of a differentiator.

dx(t )
y (t ) = .
dt

From the differentiation property,

Y ( jω ) = jωX ( jω ) ,

The frequency response of the differentiator is

Y ( jω
H ( jω ) = ) = jω .
X ( jω )

Example: Consider an integrator specified by the equation:

17/4 Yao
 ELG 3120 Signals and Systems Chapter 4

y (t ) = ∫ x(τ )dτ .
t

−∞

The impulse response of an integrator is the unit step, and therefore the frequency response of
the system:

1
H ( jω ) = + πδ (ω ) .
jω

So we have

1
Y ( jω ) = H ( jω ) X ( jω ) = X ( jω ) + πX (0)δ (ω ) ,
jω

which is consistent with the integration property.

Example: Consider the response of an LTI system with impulse response

h(t ) = e −at u (t) , a >0

to the input signal

x(t ) = e −bt u (t ) , b>0

To calculate the Fourier transforms of the two functions:

1
X ( jω ) = , and
b + jω
1
H ( jω ) = .
a + jω

Therefore,

1
Y ( jω ) = ,
(a + jω )(b + jω )

using partial fraction expansion (assuming a ≠ b ), we have

1  1 1 
Y ( jω ) = −
b − a  a + jω b + jω 


The inverse transform for each of the two terms can be written directly. Using the linearity
property, we have

18/4 Yao
 ELG 3120 Signals and Systems Chapter 4

y (t ) =
1
b−a
[
e −at u (t ) − e −bt u (t) . ]
We should note that when a = b , the above partial fraction expansion is not valid. However,
with a = b , we have

1
Y ( jω ) = ,
(a + jω )
2

1 d  1 
= j
dω  a + jω 
Considering , and
(a + jω ) 2

1
e −at u (t) ←→
F
, and
a + jω

d  1 
te −at u (t ) ←→
F
j  a + jω  ,
dω  

so we have

Y (t ) = te −at u (t ) .

4.5 The Multiplication Property

1 +∞
r (t ) = s(t ) p (t ) ←
→ R ( jω ) =
2π ∫
−∞
S ( jθ ) P ( j (ω − θ )) dθ

Multiplication of one signal by another can be thought of as one signal to scale or modulate the
amplitude of the other, and consequently, the multiplication of two signals is often referred to as
amplitude modulation.

Example: Let s (t ) be a signal whose spectrum S ( jω ) is depicted in the figure below.

19/4 Yao
 ELG 3120 Signals and Systems Chapter 4

Also consider the signal

p(t ) = cos ω 0 t ,

then

P( jω ) = πδ (ω − ω 0 ) + πδ (ω + ω 0 ) .

The spectrum of r (t ) = s(t ) p(t ) is obtained by using the multiplication property,

1 +∞
R( jω ) =
2π ∫
−∞
S ( jω ) P ( j(ω − θ ))dθ

,
1 1
= S ( jω − ω 0 ) + S ( jω + ω 0 )
2 2

which is sketched in the figure below.

From the figure we can see that the signal is preserved although the information has been shifted
to higher frequencies. This forms the basic for sinusoidal amplitude modulation systems for
communications.

Example: If we perform the following multiplication using the signal r (t ) obtained in the
preceding example and p(t ) = cos ω 0 t , that is,

g (t ) = r (t) p (t )

The spectrum of P( jω ) , R( jω ) and G ( jω ) are plotted in the figure below.

20/4 Yao
 ELG 3120 Signals and Systems Chapter 4

If we use a lowpass filter with frequency response H ( jω ) that is constant at low frequencies and
zero at high frequencies, then the output will be a scaled replica of S ( jω ) . Then the output will
be scaled version of s (t ) - the modulated signal is recovered.

21/4 Yao
 ELG 3120 Signals and Systems Chapter 4

4.6 Summary of Fourier Transform Properties and Basic Fourier Transform

Pairs

22/4 Yao
ELG 3120 Signals and Systems Chapter 4

23/4 Yao
 ELG 3120 Signals and Systems Chapter 4

4.7 System Characterized by Linear Constant-Coefficient Differential

Equations

An LTI system described by the following differential equation:

N
d k y (t) M d k x( t )
∑a k
dt k
= ∑ b k
dt k
, (4. 67)
k =0 k =0

which is commonly referred to as an Nth-order differential equation.

The frequency response of this LTI system

Y ( jω )
H ( jω ) = , (4. 68)
X ( jω )

where X ( jω ) , Y ( jω ) and H ( jω ) are the Fourier transforms of the input x(t ) , output y(t ) and
the impulse response h(t ) , respectively.

Applying Fourier transform to both sides, we have

N d k y( t )  M d k x( t ) 
F ∑ a k  = F ∑ k b , (4. 69)
 k =0 dt k   k =0 dt k 

From the linearity property, the expression can be written as

N
 d k y( t )  M  d k x( t ) 
∑ k  dt k  ∑ k  dt k  .
a F = b F (4. 70)
k =0   k =0  

From the differentiation property,

∑
M
N M
Y ( jω ) bk ( j ω ) k
∑a ( jω ) Y ( jω ) = ∑ bk ( jω ) X ( jω )
k k
⇒ H ( jω ) = = k =0
(4. 71)
X ( jω ) ∑
k
a ( jω ) k
N
k =0 k =0
k =0 k

H ( jω ) is a rational function, that is, it is a ratio of polynomials in ( jω ) .

Example: Consider a stable LTI system characterized by the differential equation

dy(t )
+ ay (t ) = x(t) , with a > 0 .
dt

The frequency response is

24/4 Yao
 ELG 3120 Signals and Systems Chapter 4

1
H ( jω ) = .
jω + a

Te impulse response of this system is then recognized as

h(t ) = e −at u (t) .

Example: Consider a stable LTI system that is characterized by the differential equation

d 2 y( t ) dy(t ) dx (t )
2
+4 + 3 y (t) = + 2 x( t ) .
dt dt dt

The frequency response of this system is

( jω ) + 2 jω + 2
H ( jω ) = = .
( jω ) + 4( jω ) + 3 ( jω + 1)( jω + 3)
2

Then, using the method of partial-fraction expansion, we find that

1/ 2 1/ 2
H ( jω ) = + .
jω + 1 jω + 3

The inverse Fourier transform of each term can be recognized as

1 −t 1
h(t ) = e u (t ) + e −3t u (t ) .
2 2

jω + 2
Example: Consider a system with frequency response of H ( jω ) = and suppose
( jω + 1)( jω + 3)
that the input to the system is

x( t ) = e − t u ( t ) ,

find the output response.

The output in the frequency domain is give as

 jω + 2  1  jω + 2
Y ( jω ) = H ( jω ) X ( jω ) =     = ,
 ( jω + 1)( jω + 3)   jω + 1  ( jω + 1) ( jω + 3) )
2

Using partial-fraction expansion, we have

25/4 Yao
 ELG 3120 Signals and Systems Chapter 4

1/ 4 1/ 2 1/ 4
Y ( jω ) = + + ,
jω + 1 ( jω + 1) 2
( jω + 3) )
By inspection, we get directly the inverse Fourier transform:

1 1 1 
h(t ) =  e −t + te −t − e −3t u (t ) .
4 2 4 

26/4 Yao
 ELG 3120 Signals and Systems Chapter 5

Chapter 5 The Discrete-Time Fourier Transform

5.0 Introduction

• There are many similarities and strong parallels in analyzing continuous-time and discrete-
time signals.
• There are also important differences. For example, the Fourier series representation of a
discrete-time periodic signal is finite series, as opposed to the infinite series representation
required for continuous-time period signal.
• In this chapter, the analysis will be carried out by taking advantage of the similarities
between continuous-time and discrete-time Fourier analysis.

5.1 Representation of Aperiodic Signals: The discrete-Time Fourier

Transform

5.1.1 Development of the Discrete-Time Fourier Transform

Consider a general sequence that is a finite duration. That is, for some integers N 1 and N 2 , x[n]
equals to zero outside the range N 1 ≤ n ≤ N 2 , as shown in the figure below.

We can construct a periodic sequence ~ x [n] using the aperiodic sequence x[n] as one period. As
~
we choose the period N to be larger, x [n] is identical to x[n] over a longer interval, as N → ∞ ,
~
x [n] = x[n] .

Based on the Fourier series representation of a periodic signal given in Eqs. (3.80) and (3.81), we
have

1/5 Yao
 ELG 3120 Signals and Systems Chapter 5

~
x [ n] = ∑a e k
jk (2π / N ) n
, (5.1)
k =< N >

ak = ∑ ~x [n]e − jk (2π / N ) n
. (5.2)
k= N

If the interval of summation is selected to include the interval N 1 ≤ n ≤ N 2 , so ~

x [n] can be
replaced by x[n] in the summation,

N2 ∞
1 1
ak =
N
∑ x[n]e − jk ( 2π / N ) n = N
∑ x[n]e − jk ( 2π / N ) n
, (5.3)
k = N1 k = −∞

Defining the function

∞
jω
X (e ) = ∑ x[n ]e
n = −∞
− j ωn
, (5.4)

So a k can be written as

1
ak = X (e jkω 0 ) , (5.5)
N

Then ~
x [n] can be expressed as

1 1
~
x [n] = ∑
k =< N > N
X (e jk ω 0 )e jk (2π / N ) n =
2π
∑ X (e jk ω0
)e jk ( 2π / N ) nω 0 . (5.6)
k =< N >

As N → ∞ ~
x [n] = x[n] , and the above expression passes to an integral,

1
x[n ] =
2π ∫ 2π
X (e jω )e jωn dω , (5.7)

The Discrete-time Fourier transform pair:

1
x[n ] =
2π ∫ 2π
X (e jω )e jωn dω , (5.8)

∞
X (e ) = jω
∑ x[n]e
n = −∞
− jω n
. (5.9)

2/5 Yao
 ELG 3120 Signals and Systems Chapter 5

Eq. (5.8) is referred to as synthesis equation, and Eq. (5.9) is referred to as analysis equation
and X (e jkω 0 ) is referred to as the spectrum of x[n] .

5.1.2 Examples of Discrete-Time Fourier Transforms

Example: Consider x[n ] = a n u[n] , a < 1. (5.10)

∞ ∞ ∞
X (e jω ) = ∑ x[n]e − jωn = ∑ a n u[n ]e − jωn = ∑ ae − jω ( ) −n
=
1
1 − ae − jω
. (5.11)
n = −∞ n= −∞ n= 0

The magnitude and phase for this example are show in the figure below, where a > 0 and a < 0
are shown in (a) and (b).

Example: x[n] = a , a < 1 .

n
(5.12)

∞ −1 ∞
X (e jω ) = ∑ a u[n ]e − jωn =
n = −∞
n
∑ a −n e − jωn + ∑ a n e − jωn
n = −∞ n =0

Let m = −n in the first summation, we obtain

∞ ∞ ∞
X (e jω ) = ∑ a u[ n]e − jωn = ∑ a m e jωm + ∑ a n e − jωn
n

n = −∞ m =1 n =0

. (5.13)
jω
ae 1 1− a 2
= jω
+ − jω
=
1 − ae 1 − ae 1 − 2 a cos ω + a 2

3/5 Yao
 ELG 3120 Signals and Systems Chapter 5

Example: Consider the rectangular pulse

1, n ≤ N1
x[n ] =  , (5.14)
0, n > N1

sin ω (N1 + 1 / 2 )
N1
X ( jω ) = ∑e
n= − N
− j ωn
=
sin (ω / 2 )
. (5.15)
1

This function is the discrete counterpart of the sic

function, which appears in the Fourier transform of
the continuous-time pulse.

The difference between these two functions is that

the discrete one is periodic (see figure) with period of 2π , whereas the sinc function is aperiodic.

5.1.3 Convergence

∞
The equation X (e jω ) = ∑ x[n ]e
n = −∞
− j ωn
converges either if x[n] is absolutely summable, that is

∑ x[n] < ∞ ,
n = −∞
(5.16)

or if the sequence has finite energy, that is

∞ 2

∑ x[n] < ∞ . (5.17)

n = −∞

4/5 Yao
 ELG 3120 Signals and Systems Chapter 5

And there is no convergence issues associated with the synthesis equation (5.8).

If we approximate an aperidic signal x[n] by an integral of complex exponentials with

frequencies taken over the interval ω ≤ W ,

1
∫
W
xˆ[ n] = X ( e jω )e jω n dω , (5.18)
2π −W

and xˆ[n ] = x[n ] for W = π . Therefore, the Gibbs phenomenon does not exist in the discrete-time
Fourier transform.

Example: the approximation of the impulse response with different values of W .

For W = π / 4, 3π / 8, π / 2, 3π / 4, 7π / 8, π , the approximations are plotted in the figure below.

)
We can see that when W = π , x[n] = x[n ] .

5/5 Yao
 ELG 3120 Signals and Systems Chapter 5

5.2 Fourier transform of Periodic Signals

For a periodic discrete-time signal,

x[n ] = e jω 0 n , (5.19)

its Fourier transform of this signal is periodic in ω with period 2π , and is given

+∞
jω
X (e ) = ∑ 2πδ (ω − ω
l = −∞
0
− 2πl ) . (5.20)

Now consider a periodic sequence x[n] with period N and with the Fourier series representation

x[n ] = ∑a e k
jk ( 2π / N ) n
. (5.21)
k =< N >

The Fourier transform is

+∞
2πk
jω
X (e ) = ∑ 2πa δ (ω −
k = −∞
k
N
). (5.22)

Example: The Fourier transform of the periodic signal

1 jω0n 1 − jω0 n 2π
x[n ] = cos ω0 n = e + e , with ω 0 = , (5.23)
2 2 3

is given as

 2π   2π 
X (e jω ) = πδ ω −  + πδ ω + , −π ≤ ω < π . (5.24)
 3   3 

6/5 Yao
 ELG 3120 Signals and Systems Chapter 5

Example: The periodic impulse train

+∞
x[n ] = ∑δ [n − kN ] .
k = −∞
(5.25)

The Fourier series coefficients for this signal can be calculated

ak = ∑ x[n ]e − jk (2π / N ) n
. (5.26)
n =< N >

Choosing the interval of summation as 0 ≤ n ≤ N − 1 , we have

1
ak = . (5.27)
N

The Fourier transform is

2π ∞
 2πk 
X (e jω ) =
N
∑ δ  ω −
k = −∞ N 
. (5.28)

7/5 Yao
 ELG 3120 Signals and Systems Chapter 5

5.3 Properties of the Discrete-Time Fourier Transform

Notations to be used

X (e jω ) = F {x[n]},

{ }
x[n ] = F −1 X (e jω ) ,

x[n ] ←→
F
X (e jω ) .

5.3.1 Periodicity of the Discrete-Time Fourier Transform

The discrete-time Fourier transform is always periodic in ω with period 2π , i.e.,

( ) ( )
X e j (ω +2π ) = X e jω . (5.29)

5.3.2 Linearity

If x1 [n] ←→
F
X 1 (e j ω ) , and x 2 [n] ←→
F
X 2 (e j ω ) ,

then

ax1[n] + bx2 [n] ←→

F
aX1 (e jω ) + bX 2 (e jω ) (5.30)

5.3.3 Time Shifting and Frequency Shifting

If x[n ] ←→
F
X (e jω ) ,

then

x[ n − n0 ] ←→
F
e − jωn0 X ( e jω ) (5.31)

and

e jω0n x[ n] ←→
F
X ( e j (ω− ω0 ) ) (5.32)

8/5 Yao
 ELG 3120 Signals and Systems Chapter 5

5.3.4 Conjugation and Conjugate Symmetry

If x[n ] ←→
F
X (e jω ) ,

then

x *[n]←→
F
X * (e− jω ) (5.33)

If x[n] is real valued, its transform X (e jω ) is conjugate symmetric. That is

X (e jω ) = X * (e− jω ) (5.34)

{ } { }
From this, it follows that Re X (e jω ) is an even function of ω and Im X (e j ω ) is an odd
function of ω . Similarly, the magnitude of X (e jω ) is an even function and the phase angle is
an odd function. Furthermore,

Ev{x[n]}←→
F
{
Re X (e jω , } (5.35)

and

Od {x[ n]}←→
F
{
j Im X (e jω . } (5.36)

5.3.5 Differencing and Accumulation

If x[n ] ←→
F
X (e jω ) ,

then

x[n] − x[n − 1] ←→

F
(
1 − e − jω X (e jω ) . ) (5.37)

For signal

n
y[ n] = ∑ x[ m] ,
m = −∞
(5.38)

its Fourier transform is given as

9/5 Yao
 ELG 3120 Signals and Systems Chapter 5

n +∞
1
∑
m = −∞
x[ m] ←→
1−e
F
− jω
X ( e ) + πX (e ) ∑ δ (ω − 2πk ) .
jω j0

m =−∞
(5.39)

The impulse train on the right-hand side reflects the dc or average value that can result from
summation.

For example, the Fourier transform of the unit step x[n ] = u[n ] can be obtained by using the
accumulation property.

We know g[n ] = δ [n] ←→

F
G(e jω ) = 1 , so

n +∞ +∞
1 1
x[n ] = ∑ g[m] ←→
m = −∞
F

(
1 − e − jω)
G (e jω
) + π G ( e j0
) ∑
k = −∞
δ (ω − 2π k
(
) =
)
1 − e − jω
+ π ∑ δ (ω − 2πk ) .
k = −∞
(5.40)

5.3.6 Time Reversal

If x[n ] ←→
F
X (e jω ) ,

then

x[−n] ←→
F
X (−e jω ) . (5.41)

5.3.7 Time Expansion

For continuous-time signal, we have

1  jω 
x( at) ←→
F
X . (5.42)
a  a 

For discrete-time signals, however, a should be an integer. Let us define a signal with k a
positive integer,

 x[n / k ], if n is a multiple of k
x( k ) [n] =  . (5.43)
0, if n is not a multiple of k

x( k ) [n] is obtained from x[n] by placing k − 1 zeros between successive values of the original
signal.

The Fourier transform of x( k ) [n] is given by

10/5 Yao
 ELG 3120 Signals and Systems Chapter 5

+∞ +∞ +∞
X ( k ) (e j ω ) = ∑ x( k ) [n ]e − jωn =
n = −∞
∑ x ( k ) [rk ]e − jωrk =
r = −∞
∑ x[r ]e
r = −∞
− j ( kω ) r
= X (e jk ω ) . (5.44)

That is,

x(k ) [n] ←→

F
X (e jkω ) . (5.45)

For k > 1 , the signal is spread out and slowed down in time, while its Fourier transform is
compressed.

Example: Consider the sequence x[n] displayed in the figure (a) below. This sequence can be
related to the simpler sequence y[n] as shown in (b).

x[n ] = y( 2 ) [n ] + 2 y (2 ) [ n − 1] ,

where

 y[n / 2], if n is even

y2 [ n ] = 
0, if n is odd

The signals y (2 ) [ n] and 2 y ( 2) [n − 1] are depicted in (c) and (d).

As can be seen from the figure below, y[n] is a rectangular pulse with N 1 = 2 , its Fourier
transform is given by

sin( 5ω / 2)
Y (e jω ) = e − j 2ω .
sin(ω / 2)

Using the time-expansion property, we then obtain

11/5 Yao
 ELG 3120 Signals and Systems Chapter 5

sin( 5ω )
y ( 2) [n] ←→
F
e − j 4ω
sin(ω )

sin( 5ω )
2 y ( 2) [n − 1] ←→
F
2e − j 5ω
sin(ω )

Combining the two, we have

 sin( 5ω ) 
X (e jω ) = e − j 4ω (1 + 2e − jω )  .
 sin(ω ) 

5.3.8 Differentiation in Frequency

If x[n ] ←→
F
X (e jω ) ,

∞
Differentiate both sides of the analysis equation X (e jω ) = ∑ x[n ]e
n = −∞
− jωn

jω
dX (e ) +∞
= ∑ − jnx[n]e − jωn .
dω n= −∞
(5.46)

The right-hand side of the Eq. (5.46) is the Fourier transform of − jnx[n] . Therefore, multiplying
both sides by j , we see that

dX (e jω )
nx[n]←→ j F

dω . (5.47)

5.3.9 Parseval’s Relation

If x[n ] ←→
F
X (e jω ) , then we have

+∞
1
∑ x[n] = ∫
2
X (e jω ) dω
2

n = −∞ 2π 2π (5.48)

12/5 Yao
 ELG 3120 Signals and Systems Chapter 5

Example: Consider the sequence x[n] whose Fourier transform X (e jω ) is depicted for
− π ≤ ω ≤ π in the figure below. Determine whether or not, in the time domain, x[n] is periodic,
real, even, and /or of finite energy.

• The periodicity in time domain implies that the Fourier transform has only impulses located
at various integer multiples of the fundamental frequency. This is not true for X (e jω ) . We
conclude that x[n] is not periodic.
• Since real-valued sequence should have a Fourier transform of even magnitude and a phase
function that is odd. This is true for X (e jω ) and ∠X (e jω ) . We conclude that x[n] is real.
• If x[n] is real and even, then its Fourier transform should be real and even. However, since
X (e jω ) = X (e jω ) e − j 2ω , X (e jω ) is not real, so we conclude that x[n] is not even.
2
• Based on the Parseval’s relation, integrating X (e j ω ) from − π to π will yield a finite
quantity. We conclude that x[n] has finite energy.

5.4 The convolution Property

If x[n] , h[n] and y[n] are the input, impulse response, and output, respectively, of an LTI
system, so that

y[ n] = x[n ] ∗ h[n] , (5.49)

then,

Y (e jω ) = X (e jω ) H (e jω ) , (5.50)

13/5 Yao
 ELG 3120 Signals and Systems Chapter 5

where X (e jω ) , H (e j ω ) and Y (e jω ) are the Fourier transforms of x[n] , h[n] and y[n] ,
respectively.

Example: Consider the discrete-time ideal lowpass filter with a frequency response H (e j ω )
illustrated in the figure below. Using − π ≤ ω ≤ π as the interval of integration in the synthesis
equation, we have

1 π
h[ n] =
2π ∫ −π
H (e j ω )e jωn dω

1 π sin ω c n
=
2π ∫
−π
e jωn dω =
πn

The frequency response of the discrete-time

ideal lowpass filter is shown in the right figure.

Example: Consider an LTI system with impulse response

h[n ] = α n u[n] , α < 1,

and suppose that the input to the system is

x[n ] = β n u[n] , β < 1.

The Fourier transforms for h[n] and x[n] are

1
H (e j ω ) = ,
1 − α e − jω

and

1
X (e jω ) = ,
1 − β e − jω

so that

1
Y (e jω ) = H (e jω ) X (e jω ) = − jω
.
(1 − α e )(1 − β e − j ω )

14/5 Yao
 ELG 3120 Signals and Systems Chapter 5

If α ≠ β , the partial fraction expansion of Y (e jω ) is given by

α β
−
A B α−β α −β
Y (e jω ) = + = + ,
(1 − α e ) (1 − β e ) (1 − α e ) (1 − β e − jω )
− jω − jω − jω

We can obtain the inverse transform by inspection:

α β
y[ n] =
α−β
α n u[n] −
α−β
β n u[n ] =
1
α−β
( )
α n +1u[n ] − ββ n +1u[n ] .

For α = β ,

1
Y (e jω ) = , which can be expressed as
(1 − α e − j ω ) 2

j jω d  1 
Y (e jω ) = e  .
− jω 
α dω  1 − α e 
Using the frequency differentiation property, we have

d  1 
nα n u[n] ←→ j
F
  ,
dω  1 − α e − j ω 

To account for the factor e jω , we use the time-shifting property to obtain

d  1 
(n + 1)α n +1u[n + 1] ←→ je jω
F
  ,
dω  1 − α e − jω 

Finally, accounting for the factor 1 / α , we have

y[ n] = (n + 1)α n u[n + 1] .

Since the factor n + 1 is zero at n = −1 , so y[n] can be expressed as

y[ n] = (n + 1)α n u[n] .

Example: Consider the system shown in the figure below. The LTI systems with frequency
response H lp (e j ω ) are ideal lowpass filters with cutoff frequency π / 4 and unity gain in the
passband.

15/5 Yao
 ELG 3120 Signals and Systems Chapter 5

• w1 [n] = (−1) n x[ n] = e j πn x[n]

⇒ W1 (e jω ) = X (e j (ω −π ) ) .

• W2 (e jω ) = H lp (e j ω ) X (e j (ω − π ) ) .

• w3 [ n] = (−1) n w2 [n ] = e jπn w2 [ n]

⇒ W3 (e j ω ) = W2 (e j (ω −π ) ) = H lp (e ( j ω − π ) ) X (e j (ω − 2π ) ) .

⇒ W3 (e j ω ) = W2 (e j (ω −π ) ) = H lp (e jω −π ) ) X (e j ω ) (Discrete-Fourier transforms are always

periodic with period of 2π ).

• W4 (e jω ) = H lp (e j ω ) ) X (e jω ) .

• [ ]
Y (e jω ) = W3 (e jω ) + W4 (e j ω ) = H lp (e ( j ω −π ) ) + H lp (e jω ) X (e j ω ) .

The overall system has a frequency response

[ ]
H lp (e jω ) = H lp (e ( jω −π ) ) + H lp (e jω ) X (e jω ) ,

which is shown in figure (b).

The filter is referred to as bandstop filter, where

the stop band is the region π / 4 < ω < 3π / 4 .

It is important to note that not every discrete-time LTI system has a frequency response. If an
LTI system is stable, then its impulse response is absolutely summable; that is,

+∞

∑ h[n] < ∞ ,
n = −∞
(5.51)

5.5 The multiplication Property

Consider y[n] equal to the product of x1 [n] and x 2 [n] , with Y (e jω ) , X 1 (e jω ) , and X 2 (e jω )
denoting the corresponding Fourier transforms. Then

16/5 Yao
 ELG 3120 Signals and Systems Chapter 5

1
y[ n] = x1 [n] x2 [n ] ←→ ∫ X 1 (e jω ) X 2 (e j ( ω −θ ) ) dθ
F
(5.52)
2π 2π

Eq. (5.52) corresponds to a periodic convolution of X 1 (e jω ) and X 2 (e jω ) , and the integral in

this equation can be evaluated over any interval of length 2π .

Example: Consider the Fourier transform of a signal x[n] which the product of two signals; that
is

x[n ] = x1 [n] x2 [n]

where

sin( 3πn / 4)
x1 [n] = , and
πn

sin(πn / 2)
x 2 [n ] = .
πn

Based on Eq. (5.52), we may write the Fourier transform of x[n]

1 π
X (e jω ) =
2π ∫
−π
X 1 (e jω ) X 2 (e j (ω −θ ) )dθ . (5.53)

Eq. (5.53) resembles aperiodic convolution, except for the fact that the integration is limited to
the interval of − π < θ < π . The equation can be converted to ordinary convolution with
integration interval − ∞ < θ < ∞ by defining

 X (e jω ) for − π < ω < π

Xˆ 1 (e j ω ) =  1
0 otherwise

Then replacing X 1 ( e jω ) in Eq. (5.53) by Xˆ 1 (e jω ) , and using the fact that Xˆ 1 (e jω ) is zero for
− π < ω < π , we see that

1 π 1 ∞
X (e jω ) =
2π ∫
−π
X 1 (e jω ) X 2 (e j (ω −θ ) )dθ =
2π ∫
−∞
Xˆ 1 (e j ω ) X 2 (e j (ω −θ ) )dθ .

Thus, X (e jω ) is 1 / 2π times the aperiodic convolution of the rectangular pulse Xˆ 1 (e jω ) and the
periodic square wave X 2 (e jω ) . The result of thus convolution is the Fourier transform X (e jω ) ,
as shown in the figure below.

17/5 Yao
 ELG 3120 Signals and Systems Chapter 5

5.6 Tables of Fourier Transform Properties and Basic Fourier Transform

Paris

18/5 Yao
ELG 3120 Signals and Systems Chapter 5

19/5 Yao
 ELG 3120 Signals and Systems Chapter 5

5.7 Duality

For continuous-time Fourier transform, we observed a symmetry or duality between the analysis
and synthesis equations. For discrete-time Fourier transform, such duality does not exist.
However, there is a duality in the discrete-time series equations. In addition, there is a duality
relationship between the discrete-time Fourier transform and the continuous-time Fourier
series.

5.7.1 Duality in the discrete-time Fourier Series

Consider the periodic sequences with period N, related through the summation

1
f [m ] = ∑ g (r )e − jr ( 2π / N ) m .
N r =< N >
(5.54)

If we let m = n and r = − k , Eq. (5.54) becomes

1
f [ n] = ∑ N
g (− r )e jr ( 2π / N ) n . (5.55)
k =< N >

Compare with the two equations below,

x[n] = ∑a e
k= N
k
jk ( 2π / N ) n
, (3.80)

1
ak =
N
∑ x[n]e
k= N
− jk ( 2π / N ) n
. (3.81)

1
we fond that g (− r ) corresponds to the sequence of Fourier series coefficients of f [n ] . That is
N

1
f [n] ←→
FS
g[− k ] . (5.56)
N

This duality implies that every property of the discrete-time Fourier series has a dual. For
example,

x[n − n 0 ] ←→
FS
a k e − jk ( 2π / N ) n 0 (5.57)
e jm ( 2π / N ) n ←→
FS
a k −m (5.58)

20/5 Yao
 ELG 3120 Signals and Systems Chapter 5

are dual.

Example: Consider the following periodic signal with a period of N = 9 .

 1 sin( 5πn / 9)
 9 sin(πn/9) , n ≠ multiple of 9
x[n ] =  (5.59)
5 , n = multiple of 9
 9

We know that a rectangular square wave has Fourier coefficients in a form much as in Eq. (5.59).
Duality suggests that the coefficients of x[n] must be in the form of a rectangular square wave.

Let g[n] be a rectangular square wave with period N = 9 ,

1, n ≤2
g[ n ] =  , (5.60)
0, 2< n ≤4

The Fourier series coefficients b k for g[n] can be given (refer to example on page 27/3)

 1 sin( 5πk / 9)
 9 sin(πk/9) , k ≠ multiple of 9
bk =  . (5.61)
5 , k = multiple of 9
 9

The Fourier analysis equation for g[n] can be written

1 2
bk = ∑ (1)e − j 2πnk / 9 . (5.62)
9 n =−2

Interchanging the names of the variable k and n and noting that x[n ] = bk , we find that

1 2
x[n ] = ∑
9 k =−2
(1)e − j 2πnk / 9 .

Let k ' = −k in the sum on the right side, we obtain

1 2
x[n ] = ∑
9 k =−2
(1)e + j 2πnk ' / 9 .

21/5 Yao
 ELG 3120 Signals and Systems Chapter 5

Finally, moving the factor 1 / 9 inside the summation, we see that the right side of the equation
has the form of the synthesis equation for x[n] . Thus, we conclude that the Fourier coefficients
for x[n] are given by

1 / 9, k ≤2
ak =  ,
0, 2< k ≤4

with period of N = 9 .

5.8 System Characterization by Linear Constant-Coefficient Difference

Equations

A general linear constant-coefficient difference equation for an LTI system with input x[n] and
output x[n] is of the form

N M

∑a
k =0
k y[n − k ] =∑ bk x[n − k ] ,
k =0
(5.63)

which is usually referred to as Nth-order difference equation.

There are two ways to determine H (e j ω ) :

• The first way is to apply an input x[n ] = e jωn to the system, and the output must be of the
form H (e j ω )e j ωn . Substituting these expressions into the Eq. (5.63), and performing some
algebra allows us to solve for H (e j ω ) .
• The second approach is to use discrete-time Fourier transform properties to solve for
H (e j ω ) .

Based on the convolution property, Eq. (5.63) can be written as

Y (e j ω )
jω
H (e ) = . (5.64)
X (e jω )

Applying the Fourier transform to both sides and using the linearity and time-shifting properties,
we obtain the expression

N M

∑ a k e − jkω Y (e jω ) = ∑ bk e − jkω X (e jω ) .
k =0 k =0
(5.65)

22/5 Yao
 ELG 3120 Signals and Systems Chapter 5

or equivalently

∑
M
jω Y (e jω ) k =0
bk e − jk ω
H (e ) = = . (5.66)
X (e jω ) ∑ a e − jkω
N
k =0 k

Example: Consider the causal LTI system that is characterized by the difference equation,

y[ n] − ay[n − 1] = x[n] , a < 1.

The frequency response of this system is

Y (e jω ) 1
H (e j ω ) = = .
X (e ) 1 − ae jω
jω

The impulse response is given by

h[n ] = a n u[n] .

Example: Consider a causal LTI system that is characterized by the difference equation

3 1
y[ n] − y[n − 1] + y[n − 2] = 2 x[n] .
4 8

1. What is the impulse response?

n
 1
2. If the input to this system is x[n ] =   u[n] , what is the system response to this input signal?
 4

The frequency response is

1 2
H (e j ω ) = = .
1 − 12 e − jω
+ 18 e − j 2 ω
(1 − 4 e )(1 − 14 e − jω )
3 − jω

After partial fraction expansion, we have

4 2
H (e j ω ) == − ,
1 − 12 e − jω
1 − 14 e − jω

The inverse Fourier transform of each term can be recognized by inspection,

n n
 1 1
h[n ] = 4  u[n] − 2  u[n ] .
 2 4

23/5 Yao
 ELG 3120 Signals and Systems Chapter 5

Using Eq. (5.64) we have

 2  1 
Y (e jω ) = H (e jω ) X (e jω ) =  3 − jω 1 − jω   1 − jω 
 (1 − 4 e )(1 − 4 e )  1 − 4 e 
.
2
= − jω
(1 − e
3
4
)(1 − e − jω )(1 − 14 e − jω )
1
4

After partial-fraction expansion, we obtain

4 2 8
Y (e jω ) = H (e jω ) X (e jω ) = − − +
1 − 14 e − jω
(
1 − 14 e − jω )
2
1 − 12 e − j ω

The inverse Fourier transform is

  1  n 1
n
 1  
2

y[ n] = − 4  − 2( n + 1)  + 8  u[n ] .
  2  4  2  

24/5 Yao
 Laplace Transform and Continuous-Time
Frequency Response

1 Definition of Laplace Transform

• Given a continuous-time signal x(t), the Laplace transform of x(t) is
defined as
Z ∞
X(s) = x(t)e−st dt. (1)
−∞

Note that X(s) is a function which takes a complex number s and

returns a complex number X(s), i.e., X(s) is a function which maps the
complex plane into the complex plane. The set of values of s for which
the integral in (1) is well-defined is called the Region of Convergence
(ROC) of X(s).

• We will see in Section 2 that the ROC is a region in the complex plane
which is bounded by lines parallel to the imaginary axis (i.e., the line
Re(s) = 0).

• Examples:
∞
1. The Laplace transform of δ(t) is −∞ δ(t)e−st dt = 1. The Laplace
R

transform integral is well defined for all values of s. Hence, the

ROC is the entire complex plane.
∞
2. The Laplace transform of u(t) is −∞ u(t)e−st dt = 0∞ e−st dt =
R R
∞
−e−st
= 1−es where by e−s∞ , we mean the limit limt→∞ e−st .
−s∞
s 0
If the real part of s is positive, then e−s∞ = 0. If the real part
of s is not positive, then e−s∞ is not well defined. Therefore, the
Laplace transform of u(t) is 1s with the ROC Re(s) > 0.
3. Consider the signal x(t) = eat u(t). The Laplace transform of x(t)
is
Z ∞ Z ∞
X(s) = eat u(t)e−st dt = eat e−st dt
−∞ 0
−(s−a)∞
Z ∞ 1−e
= e−(s−a)t dt = . (2)
0 s−a

1
 As in the previous example, we note that the real part of (s − a)
should be greater than zero for e−(s−a)∞ to be well defined. The
requirement that the real part of (s−a) should be greater than zero
is equivalent to the requirement that the real part of s should be
greater than the real part of a, i.e., that Re(s) > Re(a). Hence, the
1
Laplace transform of eat u(t) is s−a with the ROC Re(s) > Re(a).
4. Consider the signal x(t) = e−at u(−t). The Laplace transform of
x(t) is
Z ∞ Z 0
X(s) = e−at u(−t)e−st dt = e−at e−st dt
−∞ −∞
(s+a)∞
Z 0 −1 + e
= e−(s+a)t dt = . (3)
−∞ s+a

Similar to the previous examples, we conclude that, since e(s+a)∞

is 0 if Re(s + a) < 0 and not well defined otherwise, the Laplace
−1
transform of e−at u(−t) is s+a with the ROC Re(s) < −Re(a).

2 ROC of the Laplace Transform

• The observations on the ROC in the preceding examples can be gener-
alized as follows. If the signal is right sided (in other words, if there is
a time t0 before which the signal is zero, i.e., x(t) = 0 for all t < t0 ),
then the ROC of the Laplace transform of the signal is to the right
hand side of a line parallel to the imaginary axis. If the signal is left
sided (in other words, if there is a time t0 after which the signal is zero,
i.e., x(t) = 0 for all t > t0 ), then the ROC of the Laplace transform
of the signal is to the left hand side of a line parallel to the imaginary
axis. If the signal is two sided, then the ROC is the region between two
lines parallel to the imaginary axis. In particular, if the signal is causal
(which definitely means that the signal is right sided since a causal
signal takes the value zero for all t < 0), then the ROC of the Laplace
transform of the signal is to the right hand side of a line parallel to
the imaginary axis. Similarly, if the signal is anticausal, then the ROC
of the Laplace transform of the signal is to the left hand side of a line
parallel to the imaginary axis.

2
 • Given any signal x(t), the ROC of its Laplace transform is bounded by
a pole of X(s) in the sense that the boundary of the ROC has a pole
on it. If x(t) is causal, then the ROC of its Laplace transform lies to
the right hand side of all its poles and the boundary of the ROC is at
its rightmost pole.
• A system is BIBO
R∞
stable if and only if its impulse response satisfies
the property −∞ |h(t)|dt < ∞. This is equivalent to requiring that the
ROC of the Laplace transform of h(t) should include the imaginary axis.
For a causal signal, we know that the ROC of its Laplace transform
lies to the right hand side of all its poles with its boundary being at its
rightmost pole. Hence, for a causal signal, BIBO stability is equivalent
to requiring that all the poles should lie in the left half plane (i.e., the
half of the complex s plane containing complex numbers with negative
real parts).

A causal continuous-time LTI system with transfer function H(s)

is BIBO stable if and only if
all the poles of H(s) lie in the left half plane.

3 Properties of the Laplace Transform

1. Linearity of the Laplace transform: If the Laplace transform of a
signal x(t) is X(s), then the Laplace transform of αx(t) is αX(s) for
any constant α. Also, if the Laplace transforms of two signals x1 (t) and
x2 (t) are X1 (s) and X2 (s), then the Laplace transform of αx1 (t)+βx2 (t)
is αX1 (s) + βX2 (s) for any constants α and β.
2. Convolution in time domain is equivalent to multiplication in
Laplace domain: If the Laplace transforms of two signals x1 (t) and
x2 (t) are X1 (s) and X2 (s), respectively, then the Laplace transform of
the signal x1 (t) ∗ x2 (t) is X1 (s)X2 (s).
Proof: By definition, the Laplace transform of x1 (t) ∗ x2 (t) is
Z ∞ Z ∞ Z ∞ 
−st
[x1 (t) ∗ x2 (t)]e dt = x1 (τ )x2 (t − τ )dτ e−st dt
−∞ −∞ −∞

3
 Z ∞ Z ∞
= x1 (τ )x2 (t − τ )e−s(t−τ +τ ) dτ dt
−∞ −∞
Z ∞ Z ∞
= x1 (τ )e−sτ x2 (t − τ )e−s(t−τ ) dτ dt
−∞ −∞
Z ∞ Z ∞
= x1 (τ )e−sτ dτ x2 (τ1 )e−sτ1 dτ1
−∞ −∞
= X1 (s)X2 (s) (4)
where the dummy variable τ1 = t − τ was used.
3. Time shift in time domain is equivalent to modulation in
Laplace domain: If the Laplace transform of x(t) is X(s), then the
Laplace transform of x(t − t0 ) is e−st0 X(s).
Proof: By definition, the Laplace transform of x(t − t0 ) is
Z ∞ Z ∞
−st
x(t − t0 )e dt = x(t − t0 )e−s(t−t0 ) e−st0 dt
−∞ −∞
Z ∞
−st0
= e x(t1 )e−st1 dt1 = e−st0 X(s) (5)
−∞

where t1 is the dummy variable t1 = t − t0 .

Example: Since the Laplace transform of δ(t) is 1, the Laplace trans-
form of δ(t − t0 ) is e−st0 .
4. Modulation in time domain is equivalent to shift in Laplace
domain: If the Laplace transform of x(t) is X(s), then the Laplace
transform of es0 t x(t) is X(s − s0 ).
Proof: By definition, the Laplace transform of es0 t x(t) is
Z ∞ Z ∞
es0 t x(t)e−st dt = x(t)e−(s−s0 )t dt = X(s − s0 ). (6)
−∞ −∞

Example: Since the Laplace transform of u(t) is 1s , the Laplace trans-

1
form of eat u(t) is s−a .
5. If the Laplace transform of x(t) is X(s), then the Laplace transform of
x∗ (t) is X ∗ (s∗ ).
Proof: By definition, the Laplace transform of x∗ (t) is
Z ∞ Z ∞
∗ −st
[x(t)e−s t ]∗ dt
∗
x (t)e dt =
−∞ −∞
Z∞
x(t)e−s t dt]∗ = X ∗ (s∗ ).
∗
= [ (7)
−∞

4
 Example: The Laplace transform of the signal x(t) = ejωt u(t) can
1
be found to be X(s) = s−jω . Therefore, the Laplace transform of
1 1
x∗ (t) = e−jωt u(t) is X ∗ (s∗ ) = (s∗ −jω) ∗ = s+jω .

6. If the Laplace transform of x(t) is X(s), then the Laplace transform of

1
x(at) is |a| X( as ). In particular, if the Laplace transform of x(t) is X(s),
then the Laplace transform of x(−t) is X(−s).
Proof: By definition, the Laplace transform of x(at) is
Z ∞ Z ∞ at
−st
x(at)e dt = x(at)e−s a dt. (8)
−∞ −∞
at t1
∞ ∞
If a > 0, we have −∞ x(at)e−s a dt = a1 −∞ x(t1 )e−s a dt1 and if a <
R R
R∞ at t1
∞
0, we have −∞ x(at)e−s a dt = − a1 −∞ x(t1 )e−s a dt1 where t1 is the
R

dummy variable t1 = at. Therefore,

Z ∞
−s at 1 Z∞ s 1 s
x(at)e a dt = x(t1 )e− a t1 dt1 = X( ). (9)
−∞ |a| −∞ |a| a
In particular, if a is taken to be −1, we get the result that the Laplace
transform of x(−t) is X(−s).
1
Example: Since the Laplace transform of et u(t) is s−1 , the Laplace
1 1 sgn (a)
transform of eat u(at) is |a| s −1 = s−a where sgn(a) is the sign (±1)
a
of a.
7. Differentiation in time domain is equivalent to multiplication
by s in Laplace domain: If the Laplace transform of x(t) is X(s),
then the Laplace transform of dx(t)
dt
is sX(s).
dx(t)
Proof: Let the Laplace transform of dt
be denoted by Xd (s). Then,
Z ∞ dx(t) −st
Xd (s) = e dt. (10)
−∞ dt
Using integration by parts, we have
Z ∞
Xd (s) = x(t)e−st |∞
−∞ − x(t)[−se−st ]dt. (11)
−∞

Within the ROC, x(t)e−st is zero at both the limitR∞

as t → ∞ and the
limit as t → −∞ since, by definition, the integral −∞ x(t)e−st dt is well
defined in the ROC. Hence, (11) simplifies to Xd (s) = sX(s).

5
 Example: In previous examples, we found the Laplace transforms of
1 1
ejωt u(t) and e−jωt u(t) to be s−jω and s+jω , respectively. Therefore, the
Laplace transform of the signal x1 (t) = sin(ωt)u(t) is
" #
1 1 1
X1 (s) = −
2j s − jω s + jω
ω
= 2 . (12)
s + ω2

Noting that d sin(ωt)u(t)

dt
= ω cos(ωt)u(t) + sin(ωt)δ(t) = ω cos(ωt)u(t),
we find that the Laplace transform of the signal x2 (t) = cos(ωt)u(t) is
1 ω s
X2 (s) = s 2 2
= 2 . (13)
ω s +ω s + ω2
Note that
dx2 (t)
= −ω sin(ωt)u(t) + cos(ωt)δ(t)
dt
= −ω sin(ωt)u(t) + δ(t) = −ωx1 (t) + δ(t). (14)

Hence, it should be true that

sX2 (s) = −ωX1 (s) + 1. (15)

This can indeed be easily verified to be true.

Example: The current-voltage relation of a capacitor is iC = C dvdtC .
(s)
Hence, VICC(s) 1
= sC . In analogy with the relation v = iR for a resistor,
1
we say that the impedance of a capacitor is sC . Similarly, the current-
(s)
voltage relation of an inductor is vL = L dt implying that VILL(s)
di
= sL.
Hence, the impedance of an inductor is sL.

8. Integration in time domain is equivalent to multiplication by

1
s
in Laplace domain: If the Laplace transform of x(t) is X(s), then
Rt
the Laplace transform of −∞ x(τ )dτ is 1s X(s).
Proof: From the previous property, we know that if the Laplace trans-
1 (t)
form of a signal x1 (t) is X1 (s), then the Laplace transform of dxdt is
Rt dx1 (t)
sX1 (s). Defining x1 (t) = −∞ x(τ )dτ , we have x(t) = dt . Therefore,
X(s) = sX1 (s) which means that X1 (s) = 1s X(s).

6
 Alternative proof: Consider a system with the impulse response h(t) =
u(t), i.e, with the transfer function H(s) = 1s . If x(t) is the input to
Rt
this system, the output
Rt
is y(t) = u(t) ∗ x(t) = −∞ x(τ )dτ . Hence, the
1
transfer function of −∞ x(τ )dτ is H(s)X(s) = s X(s).
Example: The Laplace transform of u(t) is 1s . Hence, the Laplace
Rt
transform of tu(t) = −∞ u(τ )dτ is s12 . In general, by applying the
same procedure (n − 1) times, we find that the Laplace transform of
tn−1
the signal (n−1)! u(t) is s1n .
9. Multiplication by −t in time domain is equivalent to differen-
tiation in Laplace domain: If the Laplace transform of x(t) is X(s),
then the Laplace transform of −tx(t) is dX(s)
ds
.
Proof: By definition, the Laplace transform of −tx(t) is
Z ∞
−st
Z ∞ de−st
[−tx(t)]e dt = x(t) dt
−∞ −∞ ds
d Z∞ dX(s)
= x(t)e−st dt = . (16)
ds −∞ ds
1
Example: The Laplace transform of e−αt u(t) is s+α . Hence, the
−αt d 1 1
Laplace transform of −te u(t) is ds s+α = − (s+α)2 , i.e, the Laplace
1
transform of te−αt u(t) is (s+α) 2 . In general for any positive integer n,
tn−1 −αt 1
the Laplace transform of (n−1)!
e u(t) is (s+α)n
.

4 Inverse Laplace Transform

Given a function H(s) and the ROC, the inverse Laplace transform involves
the problem of finding the signal h(t) such that the Laplace transform of
h(t) is H(s) with the given ROC. We will see three methods for finding the
inverse Laplace transform:

1. Using partial fractions

2. Guessing an exponential solution
3. Using integration

These methods are explained below.

7
4.1 Inverse Laplace Transform Using Partial Fractions
In this method, we decompose the given function H(s) into partial fractions
and take the inverse Laplace transform of each term in the partial fraction.
Remember that the ROC of a causal signal is the right hand side of a line
parallel to the imaginary axis while the ROC of an anticausal signal is the left
hand side of a line parallel to the imaginary axis. Hence, the inverse Laplace
1
transform of s−a given the ROC Re(s) > Re(a) is eat u(t) while the inverse
1
Laplace transform of s−a given the ROC Re(s) < Re(a) is −eat u(−t). The
following examples will further illustrate the method of finding the inverse
Laplace transform by using partial fractions.
1
Example: Find the inverse Laplace transform of H(s) = (s+1)(s+2)
with the
ROC Re(s) > −1: Taking partial fractions, we have
1 1 1
= − . (17)
(s + 1)(s + 2) s+1 s+2
The poles of the two terms in the above equation are −1 and −2, respectively.
The given ROC is to the right hand side of both the lines Re(s) = −1 and
Re(s) = −2. Therefore, both terms in (17) yield causal terms. Hence,
h(t) = e−t u(t) − e−2t u(t). (18)

1
Example: Find the inverse Laplace transform of H(s) = (s+1)(s+2) with the
ROC −2 < Re(s) < −1: Taking partial fractions, we have
1 1 1
= − . (19)
(s + 1)(s + 2) s+1 s+2
The poles of the two terms in the above equation are −1 and −2, respectively.
The given ROC is to the right hand side of the line Re(s) = −2 and to the
left hand side of the line Re(s) = −1. Therefore, the first term in (19) yields
an anticausal term while the second term in (19) yields a causal term. Hence,
h(t) = −e−t u(−t) − e−2t u(t). (20)

1
Example: Find the inverse Laplace transform of H(s) = (s+1)(s+2)2
with the
ROC Re(s) > −1: Taking partial fractions, we obtain
1 1 −1 −1
2
= + + . (21)
(s + 1)(s + 2) s + 1 s + 2 (s + 2)2

8
As in the previous example, the given ROC implies that all the terms in (21)
yield causal terms. Hence,

h(t) = e−t u(t) − e−2t u(t) − te−2t u(t). (22)

4.2 Inverse Laplace Transform by Guessing an Expo-

nential Solution
From the above examples, we see that when we take partial fractions, we
get terms involving each of the poles p1 , . . . , pn of H(s) so that the inverse
Laplace transform involves terms of the form epi t u(t). If any of the poles are
repeated, then the partial fraction expansion includes additional terms. In
general, if a pole pi is repeated k times, then the partial fraction expansion
1
contains the terms (s−p i)
, . . . , (s−p1 i )k . Hence, the inverse Laplace transform
includes the terms epi t u(t), . . . , tk−1 epi t u(t). This means that we can guess
the form of the inverse Laplace transform easily by just finding the poles of
H(s). However, the guessed form of the inverse Laplace transform involves
unknown coefficients which need to be determined using the differential equa-
tion associated with the given H(s).
1
Example: Find the inverse Laplace transform of H(s) = (s+1)(s+2) with the
ROC Re(s) > −1: The poles of H(s) are −1 and −2. Because the ROC is
to the right hand side of the lines parallel to the imaginary axis and passing
through the poles, the signal h(t) must be causal. Hence, we can guess the
form of h(t) to be

h(t) = c1 e−t u(t) + c2 e−2t u(t) (23)

with c1 and c2 being coefficients to be determined. Considering a system

with transfer function H(s), we have Y (s) = X(s)H(s) if x is the input to
the system and y the output. Hence, Y (s)[s2 + 3s + 2] = X(s) and the
differential equation associated with H(s) is

ÿ(t) + 3ẏ(t) + 2y(t) = x(t). (24)

By definition, h(t) is the response of the system when δ(t) is applied as the
input signal. Hence, h(t) satisfies the equation

ḧ(t) + 3ḣ(t) + 2h(t) = δ(t). (25)

9
We have guessed the form of h(t) to be as in (23). Hence, ḣ(t) and ḧ(t) are
of the form

ḣ(t) = −c1 e−t u(t) − 2c2 e−2t u(t) + (c1 + c2 )δ(t)

ḧ(t) = c1 e−t u(t) + 4c2 e−2t u(t) − (c1 + 2c2 )δ(t) + (c1 + c2 )δ̇(t). (26)

Substituting the guessed forms of h(t), ḣ(t), and ḧ(t) into (25) and equating
the coefficients of δ(t), δ̇(t), e−t u(t), and e−2t u(t) on the two sides of the
equation, we get the following relations between c1 and c2 :

2c1 + c2 = 1
c1 + c2 = 0. (27)

Hence, c1 = 1 and c2 = −1. Therefore, from (23), h(t) = e−t u(t) − e−2t u(t).

4.3 Inverse Laplace Transform by Using Integration

The inverse Laplace transform can be evaluated as
1 Z σ+j∞
x(t) = X(s)est ds (28)
2πj σ−j∞

where σ is any real constant such that the line (σ − j∞, σ + j∞) lies in the
ROC. Note that the line (σ − j∞, σ + j∞) is parallel to the imaginary axis.
(28) implies that if X(s) is known on the line (σ − j∞, σ + j∞), then x(t)
can be found which means that the values of X(s) for all values of s can be
found. In other words, all the information content in X(s) is encapsulated
within the values of X(s) on the line (σ − j∞, σ + j∞) in the sense that
knowing X(s) on the line (σ − j∞, σ + j∞) is equivalent to knowing X(s)
throughout the complex plane.

5 Frequency Response
If the input signal is the sinusoidal signal x(t) = Aejφ ejωt u(t), then
1
Y (s) = H(s)X(s) = H(s)Aejφ . (29)
s − jω

10
y(t) can be found by taking the (causal) inverse Laplace transform of Y (s).
This can be done, for instance, by partial fractions. If the system is BIBO
stable, then the terms in the partial fraction expansion corresponding to poles
of the system yield terms that exponentially go to zero as t → ∞. Hence,
for a BIBO stable system, it can be shown that the output signal resulting
due to the input signal x(t) = Aejφ ejωt u(t) converges at steady state to the
scaled and shifted sinusoidal signal
6 H(jω) jωt
ys (t) = Aejφ |H(jω)|ej e u(t). (30)

The same conclusion can also be reached using convolution. Assuming that
the system is BIBO stable, we can neglect the homogeneous response (i.e.,
the effect of initial conditions). Hence,
Z ∞
y(t) = h(τ )Aejφ ejω(t−τ ) u(t − τ )dτ.
−∞
Z t
= Ae e jφ jωt
h(τ )e−jωτ dτ. (31)
−∞

At steady state, i.e., as t → ∞, we get

Z ∞
y(t) = Ae e jφ jωt
h(τ )e−jωτ dτ
−∞
jφ jωt 6 H(jω) jωt
= Ae e H(jω) = Aejφ |H(jω)|ej e u(t). (32)

Similarly, for a BIBO stable system, the output signal resulting due to
the input signal x(t) = A sin(ωt + φ)u(t) converges at steady state to the
scaled and shifted sinusoidal signal

ys (t) = A|H(jω)| sin(ωt + φ + 6 H(jω))u(t). (33)

11
Chapter 5

z-transform

5.1 The z-transform of sequences

Laplace transforms are used extensively to analyze continuous-time (analog) signals as well as systems that
process continuous-time signals. As you may recall, the role of the Laplace transform was to represent a
large class of continuous-time signals as a superposition of many simpler signals, sometimes called  basis
functions or  kernels . For the Laplace transform, the kernels were complex exponential signals of the form,
est , and we represented signals for which the Laplace transform existed according to the formula
˛
1
x(t) = X(s)est ds,
2πj C

where the integral is taken as a line integral along a suitable closed contour C in the complex plane. While
the integral form of the inverse Laplace transform can be a powerful tool in the analysis of continuous-time
signals and systems, we can often avoid its direct evaluation by algebraically manipulating the expression for
X(s) such that it can be represented as a sum of terms, each of which can be immediately recognized as the
Laplace transform of a known signal x(t). Then, using linearity of the Laplace transform, we can construct
the inverse transform, term by term. We can view the inverse Laplace transform as a way of constructing
x(t), piece by piece, from many (an uncountably innite number, actually) simpler signals of the form est ,
where the amount of each such signal contained in the signal x(t) is given by X(s)ds. To determine how
much of each complex exponential signal est is contained in x(t), we have the Laplace transform formula
given by
ˆ∞
X(s) = x(t)e−st dt.
−∞

For signals that are zero, for negative time, this integral can be taken over positive time, giving the one-sided,
or unilateral Laplace transform,
ˆ∞
X(s) = x(t)e−st dt.
0

For many linear time-invariant (LTI) continuous-time systems, the relationship between the input and output
signals can be expressed in terms of linear constant coecient dierential equations. The one-sided Laplace
transform can be a useful tool for solving these dierential equations. For such systems, the Laplace transform
of the input signal and that of the output signal can be expressed in terms of a  transfer function or  system
function. In fact, many of the properties, such as causality or stability, of LTI systems can be conveniently
explored by considering the system function of the continous-time system. Another helpful property of the
Laplace transform is that it maps the convolution relationship between the input and output signals in the
time domain to a conceptually simpler multiplicative relationship. In this form, LTI systems can be thought
of in terms of how they change the magnitude and phase of each of the kernel signals est individually, and
then the output of the system is given by a superposition of each of these scaled kernel signals.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

100 z-transform

For discrete-time signals, we will see that an analogous relationship can be developed between signals
and systems using the z-transform. The discrete-time complex exponential signal, zn , where z is a complex
st
number, plays a similar role to the continuous-time complex exponential signal e . We have already seen
that discrete-time signals of this form play an important role in the analysis of linear, constant coecient
dierence equations (LCCDEs), through the their aid in developing the characteristic equation and nding
solutions to homogenous LCCDEs. There is great elegance in the mathematics linking discrete-time signals
and systems through the z-transform and we could delve deeply into this theory, devoting much more time
than we will be able to here. While our treatment of the z-transform will be limited in scope, we will see
that it is an equally valuable tool for the analysis of discrete-time signals and systems. We will use the
z-transform to solve linear constant-coecient dierence equations, as well as develop the notion of discrete-
time transfer functions. We can then use it to readily compute convolution and to analyze properties of
discrete-time linear shift-invariant systems.
We note that as with the Laplace transform, the z-transform is a function of a complex variable. The
transform itself can also take on complex values. As a result, it is a complex function of a complex variable.

5.2 Unilateral (one-sided) z-transform

Now, we will begin our study of the z-transform by rst considering the one-sided, or unilateral, version of
the transform. The unilateral z-transform of a sequence {x[n]}∞
n=−∞ is given by the sum

∞
X
X(z) = x[n]z −n (5.1)
n=0

for all z such that (5.1) converges. Here, z is a complex variable and the set of values of z for which the
sum (5.1) converges is called the region of convergence (ROC) of the z-transform. The z-transform maps
sequences to functions and their associated region of convergence, such that X (z) is the z-transform of the
sequence {x[n]}∞
n=0 . When it is clear that we are discussing sequences dened for non-negative values of the
independent  time axis, or n-axis, we will write x[n] simply, and omit the brace notation {}∞
n=0 indicating
the positive n axis. The sequences for which the z-transform is dened can be real-valued, or complex valued.
Note that the summation (5.1) multiplies x[n] by a complex geometric sequence of the form z −n , such that
the series will converge whenever |x[n]| grows no faster than exponentially. The region of convergence will be
all z such that the geometrically-weighted series (5.1) converges. This region will be all values of z outside
of some circle in the complex z-plane of radius R, the  radius of convergence for the series (5.1) as depicted
in Figure (5.1).
When we call X(z) {x[n]}∞
the transform of the sequencen=0 , we imply a form of uniqueness for the z-
∞
transform. Namely, we imply that for a given sequence {x[n]}n=0 , there exists one and only one z-transform
X(z) and its associated region of convergence. Similarly, for a given z-transform X(z) , there exists one
∞
and only one sequence {x[n]}n=0 for which the series in (5.1)converges for |z| > R. The uniqueness for the
z-transform derives from properties of power series expansions of complex functions of complex variables.
Example Consider the sequence x[n] = 2n , dened for non-negative n as shown in Figure .
This discrete-time sequence has a z-transform given by

∞
X
X(z) = 2n z −n ,
n=0

which can be re-written as

∞  n
X 2
X(z) = .
n=0
z
To determine the region of convergence of this z-transform, we simply need to consider the values of z for
which the power series converges. This can be accomplished by recalling the method for summing an innite
geometric series. Recall that for a series of the form

∞
X
S= an
n=0

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.2 Unilateral (one-sided) z-transform 101

Figure 5.1: A typical region of convergence (ROC) for a unilateral z-transform. The radius of convergence,
R, is shown and the ROC is all values of z such that |z| > R.

Figure 5.2: Discrete time sequence x[n] = 2n for n ≥ 0.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

102 z-transform

where a is a complex number, we note that this is really shorthand notation for the limit

N
X
S = lim SN = lim an .
N →∞ N →∞
n=0

For the nite geometric series dening SN , we write

SN = (1 + a + a2 + . . . + aN ).

Since this is a nite series, we can multiply both sides by a to obtain

aSN = (a + a2 + . . . + aN + aN +1 ).

Subtracting, we obtain

SN − aSN = (1 − aN +1 )
SN (1 − a) = (1 − aN −1 ).

Now, if a = 1, we know that SN = N + 1. When a 6= 1, can divide both sides by (1 − a) to obtain

1 − aN +1
SN =
1−a
which is valid for all a 6= 1. Returning to the denition of S, we have that

1 − aN +1
S = lim SN = lim ,
N →∞ N →∞ 1−a
which will only be nite when |a| < 1, for which we have

1
S= .
1−a
This is a special case of the series

N2
X
S = an = (aN1 + aN1 +1 + . . . + aN2 )
n=N1

aS = (aN1 +1 + aN1 +2 + . . . + aN2 + aN2 +1 ),

leading to
S(1 − a) = (aN1 − aN2 +1 )
or
aN1 − aN2 +1
S=
1−a
so long as a 6= 1.Note that this holds even for values of a that have magnitude greater than one. When
N2 = ∞ , we may consider
aN1 − aN2 +1 aN1
lim S = lim = ,
N2 →∞ N2 →∞ 1−a 1−a
1
so long as |a| < 1. When N1 = 0, this takes the form S= 1−a seen above. To summarize, we have seen that

N2
X aN1 − aN2 +1
an = , for a 6= 1 (5.2)
1−a
n=N1

and
∞
X aN1
an = , for |a| < 1 . (5.3)
1−a
n=N1

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.2 Unilateral (one-sided) z-transform 103

Now returning to our example, for x[n] = 2n , n ≥ 0, let us nd the ROC for X(z), the z-transform of x[n].
Is z=1 in the ROC of X(z)? Is z = 3 in the ROC? First consider z = 1.
∞
X
X(1) = 2n 1−n
n=0
X∞
= 2n ,
n=0

which clearly diverges. Therefore, z=1 is not in the ROC. Now consider z = 3.
∞
X
Z(3) = 2n 3−n
n=0
∞  n
X 2
=
n=0
3
1
= 2
1− 3
= 3.

Thus, X(z) is well-dened at z=3 and therefore z = 3 is a point in the ROC of X(z).
In this example, we saw that a larger value of z was in the ROC, whereas a smaller value was not. It
should not be a surprise that larger values of z are more likely to be in the ROC. Why so? Because, in the
denition of the z-transform, z is raised to a negative power and multiplied by the sequence x[n]. Therefore,
the z-transform is essentially a sum of the signal x[n] multiplied by either a damped or a growing complex
exponential signal z −n . Thus, larger values of z oer greater likelihood for convergence of the z-transform
sum, since these correspond to more rapidly decaying exponential signals. In general, X(z) converges for
all z that are large enough, that is, when z is suciently large, that the signal x[n]z =n becomes summable.
Specically, X(z) converges for all z such that |z| > R (for some R). Thus, the ROC of X(z) includes all
points z lying outside a circle of radius R, as illustrated in Figure 5.1. To discover the value of R for a given
sequence, we need only consider the convergence test that we need to apply when we try to compute the
z-transform sum.
For our example, we have
∞  n
X 2
,
n=0
z
which, when applying the formula (5.3)for a geometric series, yields

∞  n
X 2
X(z) = .
n=0
z
1 2
= 2 ,

<1
1− z z
z
= , |z| > 2,
z−2
that is the ROC of X(z) is |z| > 2. We can look at a more general example, such as that considered next.
Example
Consider the sequence x[n] = an , for n ≥ 0, where a is a possibly complex constant. To determine X(z),
we consider the sum

∞
X
X(z) = an z −n
n=0
∞ 
X a n
= ,
n=0
z

©A.C Singer and D.C. Munson, Jr. March 12, 2011

104 z-transform

which for |z| > |a| converges to

z
X(z) = .
z−a
Note that
a |a|
<1⇔ < 1 ⇔ |a| < |z| ⇔ |z| > |a|.
z |z|
This, we have
z
X(z) = , |z| > |a|.
z−a
What happens for |z| < |a|? Although the algebraic expression z/(z − a) can be evaluated for any value of
z 6= a, this is most certainly not the z-transform, since we know that the innite sum dening X(z) does
not converge for such values of z. Therefore, X(z) is dened only on its ROC and is not dened elsewhere.
Hence, when we mention the z-transform of a sequence, we need to not only provide an expression for X(z),
but to also dene the values of z for which this expression holds, i.e. the ROC.
Linearity
We can also use some elementary calculus to extend some of the relationships developed thus far. First,
let us show that the z-transform is linear, that is if X1 (z) is the z-transform for the sequence x1 [n] and
X2 (z) is the z-transform for the signal x2 [n] , then the signal x3 [n] = ax1 [n] + bx2 [n] is given by X3 (z) =
aX2 (z) + bX1 (z). This superposition property can be shown directly from the denition of the z-transform,
∞
X
X3 (z) = (ax1 [n] + bx2 [n])z −n
n=0
X∞ ∞
X
= ax1 [n]z −n + bx2 [n]z −n
n=0 n=0
X∞ X∞
= a x1 [n]z −n + b x2 [n]z −n
n=0 n=0
= aX1 (z) + bX2 (z).

Example
Now, to determine the z-transform of a sequence of the form x[n] = nan , we can use linearity of the
n
transform to obtain the desired result. We know that for the sequence x[n] = a we have

∞
X z
X(z) = an z −n = , |z| > |a|
n=0
z−a

and that if we dierentiate this expression with respect to z we have

∞
!  
d d X
n −n d z
X(z) = a z = , |z| > |a|
dz dz n=0
dz z−a
∞
X −a
= − nan z −n−1 = , |z| > |a|.
n=0
(z − a)2

From this expression, we can multiply by −z and obtain

∞
d X az
−z X(z) = nan z −n = , |z| > |a|.
dz n=0
(z − a)2

That is, we have the relation

∞
X az
nan z −n = , |z| > |a|.
n=0
(z − a)2

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.2 Unilateral (one-sided) z-transform 105

In a similar manner, we can obtain the more general result

 
d
nx[n] ⇔ −z X(z) ,
dz

for X(z) the z-transform of x[n]. We can continue to dierentiate to obtain the relation

1 a2 z
n(n − 1)x[n] ⇔ , |z| > |a|,
2 (z − a)3

and m-fold dierentiation leads to the relation

1 am z
n(n − 1) · · · (n − m + 1)an ⇔ , |z| > |a|.
m! (z − a)m+1

Example
We can use linearity of the z-transform to compute the z-transform of trigonometric functions, such as
x[n] = cos(ωn), for n ≥ 0. Note that rather than using x[n] = cos(ωn)u[n], we instead use the notation
n ≥ 0, since the unilateral z-transform for both sequences would be the same. From Euler's relation, we
have

∞
X
X(z) = cos(ωn)z −n
n=0
∞
X 1 jωn
= (e + e−jωn )z −n
n=0
2
∞ ∞
1 X jω −1 n 1 X −jω −1 n
= (e z ) + (e z )
2 n=0 2 n=0
1 1 1 1
= + , |z| > |ejω | = 1
2 1 − ejω z −1 2 1 − e−jω z −1
1 z 1 z
= + , |z| > 1
2z−e jω 2 z − e−jω
z(z − e−jω ) z(z − ejω )
 
1
= + , |z| > 1
2 z 2 − z(ejω + e−jω ) + 1 z 2 − z(ejω + e−jω ) + 1
z 2 − z cos(ω)
= , |z| > 1.
z 2 − 2z cos(ω) + 1

We could have shortened the derivation by using our knowledge that cos(ωn) is a sum of two complex
exponentials of the form an where a = e±jω and then use linearity together with our knowledge of the
z-transform for an . Let us now use this approach to nd the z-transform for x[n] = sin(ωn). We have that

1 jωn
− e−jωn


x[n] = e
2j
1 jω
n 1 −jω
n
= e − e
2j 2j
to which we can apply transform pairs we already know. From the z-transform of a single complex exponen-
tial, we have

1 z 1 z
X(z) = − , |z| > 1
2j z − ejω 2j z − e−jω
1 z(z − e−jω ) − z(z − ejω )
= , |z| > 1
2j z 2 − 2z cos(ω) + 1
z sin(ω)
= , |z| > 1.
z 2 − 2z cos(ω) + 1

©A.C Singer and D.C. Munson, Jr. March 12, 2011

106 z-transform

Example
From the denition of the z-transform, it should be clear that the unit sample function, i.e. the discrete-
time impulse, has a z-transform
δ[n] ⇔ 1.
Similarly, directly from the denition of the z-transform, a discrete-time impulse at n = k, i.e. δ[n − k] has
the z-transform
δ[n − k] ⇔ z −k ,
so long as k ≥ 0. Note that if k < 0, then the summation for the unilateral z-transform will never see the
only non-zero term, and hence the z-transform will be zero for δ[n + k] for k > 0.
Another sequence for which we can apply knowledge of an existing transform is the unit step, u[n]. Note
that for n ≥ 0, the unit step is a complex exponential sequence of the form an for the specic case a = 1.
As a result, we know that the z-transform for u[n] is given by

z
u[n] ⇔ , |z| > 1.
z−1

5.3 Properties of the unilateral z-transform

We will discuss a few properties of the unilateral z-transform. To facilitate this discussion, we will use the
following operator notation for the z-transform, Z(y[n]) , Y (z). The rst property has already been shown,
and is that of linearity.

5.3.1 Linearity
The unilateral z-transform is a linear operation, i.e. it satises superposition. This has been shown previously,
and we have that
Z(ay1 [n] + by2 [n]) = aY1 (z) + bY2 (z).
This is readily shown from the denition of the z-transform, i.e.
∞
X ∞
X ∞
X
(ay1 [n] + by2 [n])z −n = a y1 [n] + b y2 [n]
n=0 n=0 n=0
= aY1 (z) + bY2 (z).

5.3.2 Delay Property #1

The next property is the rst delay property, that is, when a sequence is delayed by a positive amount. If a
sequence is delayed by k samples, then we have that

Z(y[n − k]u[n − k]) = z −k Y (z).

In words, this property states that truncating a sequence at the origin, and then shifting to the right by a
positive integer k, is equivalent to multiplying the z-transform of the un-shifted sequence by z −k . This can
be proven from the detion of the z-transform. We have that
∞
X
Z(y[n − k]u[n − k]) = y[n − k]u[n − k]z −n
n=0
X∞
= y[n − k]z −n
n=k
X∞
= y[m]z −(m+k)
m=0
−k
= z Y (z),
where the second line follows from u[n − k] being zero for n<k and the third line follows from making the
substitution m = n − k.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.3 Properties of the unilateral z-transform 107

5.3.3 Delay Property #2

For cases where y[−1], y[−2], . . . , y[−k] are known or dened (k > 0), we have the following property. Here,
the sequence y[n] is not truncated at the origin, prior to shifting.
" k
#
X
−k m
Z(y[n − k]) = z Y (z) + y[−m]z .
m=1

This can be shown from linearity and delay property #1. Specically, we note that for n ≥ 0, we have that

k
X
y[n − k] = y[n − k]u[n − k] + y[−m]δ[n − k + m]
m=1

by simply adding back into the sequence the new values that shift into the region n ≥ 0 from the left.
We now can use linearity together with delay property #1 and the z-transform for a shifted discrete-time
impulse to obtain

k
X
Z(y[n − k]) = z −k Y (z) + y[−m]z −(k−m)
m=1
" k
#
X
−k m
= z Y (z) + y[−m]z .
m=1

5.3.4 Advance Property

The following advance property can also be used in the solution of dierence equations with initial conditions.
We have that " #
k−1
X
k −m
Z(y[n + k]u[n]) = z Y (z) − y[m]z .
m=0
This property is also readily shown by noting that

∞
X
Z(y[n + k]u[n]) = y[n + k]z −n
n=0
∞
X
= zk y[n + k]z −(n+k)
n=0
X∞
= zk y[m]z −m
m=k
∞ k−1
!
X X
k −m −m
= z y[m]z − y[m]z
m=0 m=0
k−1
!
X
k −m
= z Y (z) − y[m]z .
m=0

5.3.5 Convolution
One of the useful properties of the z-transform is that it maps convolution in the time domain into multipli-
cation in the z-transform domain. We will show this here for the unilateral z-tranform and sequences that
are only nonzero for n ≥ 0 and revisit the more general case when we explore the two-sided z-transform.
Specally, we assume that x[n] = h[n] = 0, n < 0 and consider the convolution
∞
X
y[n] = h[m]x[n − m].
m=−∞

©A.C Singer and D.C. Munson, Jr. March 12, 2011

108 z-transform

Taking the z-transform of both sides, we have

∞
X
Y (z) = y[n]z −n
n=0
X∞ ∞
X
= h[m]x[n − m]z −n
n=0 m=−∞
X∞ ∞
X
= h[m] x[n − m]z −n
m=−∞ n=0
∞
X
= h[m]X(z)z −m
m=−∞
∞
X
= X(z) h[m]z −m
m=−∞
= X(z)H(z),

where in the third line we used the delay property and that both sequences were zero for n < 0. When the
sequences x[n] and h[n] are not both zero for n < 0, then multiplication of one-sided z-transforms cam be
shown to be equivalent to convolution of the sequences x[n]u[n] and h[n]u[n], i.e.

∞
X n
X
x[n − k]y[n − k]h[k]u[k] = x[n − k]h[k] ←→ X(z)H(z),
k=−∞ k=0

where X(z) and H(z) are the one-sided z-transforms of the sequences x[n] and h[n].

5.3.6 Inverse unilateral z-transform

One method that can be used to solve dierence equations, is to take the z-transform of both sides of the
dierence equation, and solve the resulting algebraic equation for Y (z), and then nd the inverse transform
to obtain y[n]. A formula for the inverse unilateral z-transform can be written


1
y[n] = Y (z)z n−1 dz
2πj

which is an integral taken over a closed contour in a counter clockwise direction in the region of converge
of Y (z), as shown in Figure . Other inversion methods exist if Y (z) is a rational function (i.e., a ratio of
polynomials), e.g.,

b0 + b1 z + . . . + bM z M
Y (z) = .
a0 + a1 z + . . . + aN z N

Direct long division

A straightforward, but not entirely practical method, since it does not produce a closed-form expression
for y[n], is to employ long-division of the polynomials directly. This is a simple method for obtaining a power-
series expansion for Y (z) from the rational expression, and then from the denition of the z-transform, the
terms of the sequence can be identied one at a time.

Example
©A.C Singer and D.C. Munson, Jr. March 12, 2011
5.3 Properties of the unilateral z-transform 109

Figure 5.3: Contour integral for taking an inverse z-transform.

For the expression Y (z) = z/(z − a), we have that

z
Y (z) =
z−a
a2
1 + az + z2
z − a)z
z−a
z 0+a+0
=
z−a 0 + a − az
2

2
0 + 0 + az
2
a3
0 + 0 + az − z2

Note that from the above series expansion, together with the denition of the unilateral z-transform, i.e.
Y (z) = y[0] + y[1]z −1 + y[2]z −2 + · · · , we can immediately identify all of the terms of the sequence y[n].
That is we have that

Y (z) = 1 + az −1 + a2 z −2 + a3 z −3 + · · ·
= y[0] + y[1]z −1 + y[2]z −2 + y[3]z −3 + · · · ,

from which we may infer that y[n] = an , n ≥ 0.

5.3.7 z-transform properties

A short table of z-transform properties is given in Table (5.1) . These can be proven either directly from the
denition of the z-transform, or through application of other known properties.

5.3.8 Table of unilateral z-transform pairs

A short table of unilateral z-transforms is given in Table (5.2) below. These can also be derived directly from
the denition of the unilateral z-transform, or through application of the theorems listed in Table (5.1).

©A.C Singer and D.C. Munson, Jr. March 12, 2011

110 z-transform

Superposition ax1 [n] + bx2 [n] ⇔ aX1 (z) + bX2 (z)

Advance x[n + 1]u[n] ⇔ z(X(z) − x[0])
−1
Modulation an x[n] ⇔ X(a z)
dX(z)
Multiplication by n nx[n] ⇔ −z dz
n
X
Convolution x[k]y[n − k] ⇔ X(z)Y (z)
k=0
X∞
Convolution when x[n] = y[n] = 0, n < 0 x[k]y[n − k] ⇔ X(z)Y (z)
k=−∞ " #
k−1
X
−m
Advance by k y[n + k]u[n] ⇔ z k Y (z) − y[m]z
m=0
k
Delay property #1 y[n − k]u[k] ⇔ " z Y (z) #
Xk
Delay property #2 y[n − k] ⇔ z −k Y (z) + y[−m]z m

m=1

Table 5.1: Table of unilateral z-transform properties.

∞
X
x[n] ⇔ X(z) = x[n]z −n ROCX
( n=0
1, n = 0
δ[n] = ⇔ 1 all z
6 0
0, n =
(
z −k , k ≥ 0
δ[n − k] ⇔ z 6= 1
0, k<0
z
an ⇔ |z| > |a|
z−a
az
nan ⇔ |z| > |a|
(z − a)2
az sin(ω)
an sin(ωn) ⇔ |z| > |a|
z 2 − 2az cos(ω) + a2
1 − az cos(ω)
an cos(ωn) ⇔ |z| > |a|
( z 2 − 2az cos(ω) + a2
1, n = 0 z
u[n] = ⇔ |z| > 1
0, n 6= 0 z−1
z
1 ⇔ |z| > 1
z−1

Table 5.2: Table of unilateral z-transform pairs.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.4 Inverse z-transform by partial fraction expansion 111

5.4 Inverse z-transform by partial fraction expansion

One method for nding the inverse of a unilateral (one-sided) z-transform is to recognize the transform of
interest as the transform of a signal whose z-transform you already know, or have access to via a lookup
table, such as that found at the end of the last chapter. For example, if you know that the transform of the
unit-sample (or discrete-time impulse) is X(z) = 1, then given a transform of the form 1 X2 (z) = 2 + z =1 ,
you might use the linearity property of the z-transform together with the delay property to identify x[n] =
2δ[n] + δ[n=1]. This method is sometimes referred to as the  table lookup method . We can generalize
this idea to nd the inverse transform of more elaborate functions by learning how to decompose complex
expressions into a linear combination of terms, each of which we might be able to identify their inverse by
inspection. This simply amounts to using linearity to break a complex transform into a sum of simpler terms,
and then using a lookup table to nd the inverse of each of the terms independently. The overall inverse
transform would then be the sum of the inverses of each of the simpler terms, exploiting the linearity of the
transform.
The inverse transform method we will describe will work well in the case when X(z) is a rational function,
that is, when it can be expressed as a ratio of nite order polynomials in z . The method is based on the
notion that every rational function can be expanded in terms of partial fractions. If the rational function
X(z) is proper, that is, the degree of the numerator polynomial is less than the degree of the denominator
polynomial, and if the roots of the denominator polynomial are distinct, then we can factor X(z) in the form
b0 + b1 z + · · · bM z M b0 + b1 z + · · · bM z M
X(z) = = ,
a0 + a1 z + · · · + aN z N (z − r1 )(z − r2 ) · · · (z − rN )
where, here, X(z) is proper if M <N , and the roots of denominator polynomial are rk N
k=1 . When the rk
are distinct (or  simple ), then, we can write

N
X Ak
X(z) = ,
z − zk
k=1

where the constants Ak are called the residues of X(z) . In this form, we can use a simple method to nd
the residues when all of the roots are distinct. We see that they can be obtained by the formula

Ak = (z − rk )X(z) |z=rk ,
since the term (z rk ) makes each term in the sum become zero when evaluated at z = rk , except for the
one term in the sum that had(z rk ) in the denominator. This term is has Ak in the numerator, and hence
yields the formula above.
Once we have expanded X(z) in this form, we can then read o the inverse transform as

N N N
X Ak X Ak z X
X(z) = = z −1 ⇔ x[n] = Ak rkn−1 u[n − 1],
z − zk z − rk
k=1 k=1 k=1

once again using a combination of the linearity property of the unilateral z-transform and the delay property.
We can see how this works in practice by looking at an example.
Example
We can use this approach to nd the inverse transform for the following unilateral z-transform:

z−1
Y (z) = .
(z − 2)(z − 3)
Now, we wish to nd the sequence y[n], for n ≥ 0. We have that

z−1 A1 A2
Y (z) = = + ,
(z − 2)(z − 3) z−2 z−3
so that when we multiply Y (z) by z−2 we obtain

(z − 2)A2
(z − 2)Y (z) = A1 + .
z−3

©A.C Singer and D.C. Munson, Jr. March 12, 2011

112 z-transform

Now, setting z = 2,we have

z − 2Y (z)|z=2 = A1 ,
since the second term on the right hand side becomes zero. We then nd that

z−1 1
A1 = = = −1.
z−3 z=2 −1
Similarly we nd that
z−1 2
A2 = = = 2.
z−2 z=3 1
Putting these together yeilds that

−1 2
Y (z) = +
z−2 z−3
   
−1 z −1 z
= −z + 2z .
z−2 z−3
From the table of unilateral z-transform pairs, we have that

z
an ⇔ ,
z−a
and applying Delay Property #1, we ahve that

z
an−1 u[n − 1] ⇔ z −1 .
z−a
From the linearity of the z-transform, we can now invert each of the terms individually, and then put them
together to obtain
y[n] = −(2)n−1 u[n − 1] + 2(3)n−1 u[n − 1].
If we prefer, we can re-write this as
(
− 21 (2)n + 32 3n , n≥1
y[n] =
0, n=0.

We do not evaluate y[n] for values of n < 0, since the unilateral z-transform does not tell us anything about
this region. In this example, we needed to apply both linearity and Delay Property #1. We can avoid the
need to apply the delay property to each term, by expanding z =1 Y (z) in a PFE as

Y (z) A1 A2 A3
= + + .
z z z−2 z−3
Then we can obtain
A2 z A3 z
Y (z) = A1 + + ,
z−2 z−3
and each of the terms in this expansion can be inverted directly, without the need for the delay property.
Working out the details for this example, we have

Y (z) z−1
=
z z(z − 2)(z − 3)
A1 A2 A3
= + + ,
z z−2 z−3
and that

z−1 −1 1
A1 = = =− ,
(z − 2)(z − 3) z=0 (−2)(−3) 6
z−1 1 1
A1 = = =− ,
z(z − 3) z=2 (2)(−1) 2
z−1 2 2
A3 = = = .
z(z − 2) z=3 (3)(1) 3

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.4 Inverse z-transform by partial fraction expansion 113

Putting these together, yields

Y (z) − 16 − 12 2
= + + 3
z z z−2 z−3
1
−6z −1z 2
z
Y (z) = + 2 + 3 .
z z−2 z−3
We can again invert each term, term by term, to obtain

1 1 2
y[n] = − δ[n] − (2)n u[n] + (3)n u[n].
6 2 3
Here we have identied that the inverse transform of a constant is a discrete-time impulse. This can be
obtained either from the table of transforms, or by noting that if a z-transform is constant, say X(z) = C ,
then we have that
X(z) = x[0] + x[1]z −1 + x[2]z −2 + x[3]z −2 + · · ·
and we see that the only way that X(z) can be a constant (i.e. the only power of z in the expression is z0)
is for x[0] = C, i.e. we have that
1 1
X(z) = − ⇔ x[n] = − δ[n].
6 6
Putting all of the terms together yields,
(
− 21 (2)n + 23 (3)n , n ≥ 1
y[n] =
− 61 − 12 + 23 , n=0
(
− 21 (2)n + 23 (3)n , n ≥ 1
=
0, n = 0,

as we had before. In this example, the PFE for z =1 Y (z) was more complicated (involved one more term)
than the PFE for Y (z) . In many cases this extra complication does not arise. If the numerator of Y (z)
2
contains a power of z (say z or z ), then the z in the denominator of z
=1 Y (z) is cancelled, in which case the
PFE for z
= 1
Y (z) has exactly the same form as the PFE of Y (z).
If the ri are not distinct, we will need to modify the partial fraction expansion slightly. Suppose rj is
a root that is repeated q times. We then must replace the single term corresponding to rj with a set of q
terms, one for each occurrence of the root, where the denominator is raised to each power, starting from the
rst power up to the q th power, i.e. we replace

q
Ak X B`
⇒
(z − rk ) (z − rk )`
`=1

in the partial fraction expansion, where the new constants satisfy

dq−`
 
1 q
B` = (z − rk ) Y (z) .
(q − `)! dz q−` z=rk

While it is important to know that this formula exists, in practice, the form of the expansion is more
important than the explicit formula for determination of the constants. For example, you can determine the
constants by simply matching terms in the expansion as shown in the next example.
Example
Determine the partial fraction expansion of the z-transform

z
Y (z) = .
(z − 1)(z − 3)2
To accomplish this, we need only know the form of the expansion, and not dwell on the formula for the
constants of the repeated roots. First, we obtain

Y (z) 1 A1 A2 A3
= = + +
z (z − 1)(z − 3)2 (z − 1) (z − 3) (z − 3)2

©A.C Singer and D.C. Munson, Jr. March 12, 2011

114 z-transform

as the form of the partial fraction expansion. We can now obtain the rst term directly, using the non-
repeated roots formula
1 1
A1 = = ,
(z − 3)2 z=1 4
to get started. Now, we nd A3 before we nd A2 . In general, if we nd the coecient over the highest
power denominator rst, the resulting algebra will be simplied. By multiplying both sides of the PFE by
(z 3)2 we obtain
(z − 3)2 Y (z) 1 A1 (z − 3)2
= = + A2 (z − 3) + A3 .
z (z − 1) (z − 1)
Setting z = 3, we have
1 1
A3 = = .
z−1 z=3 2
There are a few ways to determine A2 . One is to rst dierentiate the expression (z − 3)2 Y (z)/z with
respect to z, which yields

−1 2A1 (z − 3)(z − 1) − A1 (z − 3)2

2
= + A2 ,
(z − 1) (z − 1)2
which upong setting z = 3, yields
1
− = A2 .
4
Another way to nd A2 would be to simply ll in the known constants, yielding

1 1
1 4 A2 2
= + +
(z − 1)(z − 3)2 (z − 1) (z − 3) (z − 3)2
1 2 1
4 (z − 3) + 2 (z − 1) + A2 (z − 1)(z − 3)
= .
(z − 1)(z − 3)2
Now, the numerators must match, so we must have

1 1
1= (z − 3)2 + (z − 1) + A2 (z − 1)(z − 3),
4 2
which can be easily solved for A2 . For example, both sides must have the same coecient to the term
z 2 ,which, on the left hand side is zero, and on the right hand side is

1
0= + A2 ,
4
which yields that
1
A2 = −
4
as before. Substituting these values into the original PFE yields

1 1 1
4z 4z 2z
Y (z) = − + .
(z − 1) (z − 3) (z − 3)2
The rst two terms are easy to invert from our table of known transforms. For the third term, we recall that

az
nan ⇔ .
(z − a)2
Therefore we have that
 
1 n 1 n 1 1
y[n] = (1) − (3) + n(3)n , n ≥ 0
4 4 2 3
1 1 n 1
= − (3) + n(3)n , n ≥ 0.
4 4 6

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.4 Inverse z-transform by partial fraction expansion 115

Let us consider another example.

Example
Given the unilateral z-transform of the seqeunce y[n] is given by

2z 3 + z 2 − z + 4
Y (z) = ,
(z − 2)3

nd y[n]. Recall that for a  strictly proper rational function, we require that the degree of the numerator
polynomial be strictly less than the degree of the denominator polynomial. This condition is necessary for
us to use the form of the partial fraction expansion we have considered thus far. We can use the PFE form
if we choose to expand Y (z)/z in PFE, since this will be a strictly proper rational function. We begin with

Y (z) 2z 3 + z 2 − z + 4
=
z z(z − 2)3
A1 A2 A3 A4
= + + +
z (z − 2) (z − 2)2 (z − 2)3

and immediately note that

2z 3 + z 2 − z + 4 4 1
A1 = = =− .
(z − 2)3 z=0 −8 2
Now, we again nd the coecient of repeated-root term with highest power denominator rst. Mutliplying
Y (z)/z by (z 2)3 , we obtain

2z 3 + z 2 − z + 4 (z − 2)3
=− + A2 (z − 2)2 + A3 (z − 2) + A4 ,
z 2z
which when evaluated for z = 2, yields

16 + 4 − 2 + 4
= A4
2
11 = A4 .

Now putting the PFE into a common demonimator and setting the numerators equal yields,

1
2z 3 + z 2 − z + 4 = − (z − 2)3 + A2 z(z − 2)2 + A3 z(z − 2) + 11z.
2
We can now match terms with corresponding powers of z to obtain

1
2z 3 = − z 3 + A2
2
5
= A2 ,
2
and

1 2 5
z2 = 6z + (−4)z 2 + A3 z 2
2 2
8 = A3 .

Putting all of the terms together, yields,

5
1 z 8z 11z
Y (z) = − + 2 + 2
+ .
2 z − 2 (z − 2) (z − 2)3

Now we can invert each of the terms, one at a time, to yield,

1 5 11
y[n] = − δ[n] + (2)n + 4n(2)n + (n − 1)n(2)n−2 , n ≥ 0,
2 2 2

©A.C Singer and D.C. Munson, Jr. March 12, 2011

116 z-transform

where for the last term, we used the transform pair

1 a2 z
n(n − 1)an ⇔ .
2 (z − a)3
We could combine all of the results to obtain
(
2, n=0
y[n] = 1 2

n
8 11n + 21n + 20 (2) , n ≥ 1.

5.5 Dierence equations and the z-transform

Just as the Laplace transform was used to aid in the solution of linear dierential equations, the z-transform
can be used to aid in the solution of linear dierence equations. Recall that linear, constant coecient
dierential equations could be converted into algebraic equations by transforming the signals in the equation
using the Laplace transform. Derivatives could be mapped into functions of the Laplace transform variable s,
through the derivative law for Laplace transforms. Similarly, delayed versions of a sequence can be mapped
into algebraic functions of z, using one of the delay rules for z-transforms.
In the case of continuous-time linear systems described by dierential equations, in order to nd the
response of such a linear system to an particular input, the dierential equations needed to be solved, using
either time-domain or Laplace transform methods. For an N th-order dierential equation, in general N
conditions on the output were needed in order to specify the output in response to a given input. Similarly,
for linear dierence equations of N th-order, N pieces of information are needed to nd the output for a
given input. Unlike the continuous-time case, dierence equations can often be simply iterated forward in
time if these N conditions are consecutive. That is, given y[−N + 1], ...y[−1], then re-writing

N
X M
X
ak y[n − k] = bk x[n − k]
k=0 k=0

in the form

N
X M
X
a0 y[0] = − ak y[n − k] = bk x[n − k],
k=1 k=0
from which y[0] could be found. Iterating this process forward could nd each value of the output without
ever explicitly obtaining a general expression for y[n].
In this chapter we will explore the z-transform for the explicit solution of linear constant coecient
dierence equations. The properties of the z-transform that we have developed can be used to map the
dierence equations describing the relationship between the input and the output, into a simple set of linear
algebraic equations involving the z-transforms of the input and output sequences. By solving the resulting
algebraic equations for the z-transform of the output, we can then use the methods we've developed for
inverting the transform to obtain an explicit expression for the output. We begin with an example.
Example
We revisit this simple linear, homogeneous dierence equation, now using the unilateral ztransform.
Again consider the dierence equation

y[n] − 3y[n − 1] = 0, n ≥ 0, y[−1] = 2

Taking unilateral z-transform of both sides, and using the delay property, we obtain

Y (z) − 3z −1 [Y (z) + zy[−1]] = 0

Y (z)[1 − 3z −1 ] = 6,
which can be solved for Y (z) directly, yielding

6z
Y (z) = ,
z−3
n
y[n] = 6(3) u[n].

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.5 Dierence equations and the z-transform 117

Another, slightly more involved example, repeats another example as well.

Example
Consider the following homogenous, linear constant coecient dierence equation, dened for nonnegative
n and with initial conditions shown

y[n] + 4y[n-1] + 4y[n − 2] = 0, n≥0, y[-1] = y[−2] = 1.

Taking the z-transform of both sides, again using the delay property and including the initial conditions, we
obtain

Y (z) + 4z −1 [Y (z) + zy[−1]] + 4z −2 Y (z) + zy[−1] + z 2 y[−2] = 0

 

Y (z) 1 + 4z −1 + 4z −2 = −4y[−1] − 4z −1 y[−1] − 4y[−2]

 

−8 − 4z −1
Y (z) =
1 + 4z −1 + 4z −2
−8z 2 − 4z
= .
z 2 + 4z + 4
This is not in strictly proper rational form, i.e. the degree of the numerator is not strictly less than that of
the denominator, however when we expand z −1 Y (z), we have

Y (z) −8z − 4 −8z − 4

= =
z z2 + 4z + 4 (z + 2)2
A1 A2
= + .
(z + 2) (z + 2)2
Since we have repeated roots, we rst seek the coecient of the highest order root, A2 . By cross multiplying,
we obtain
−8z − 4 = A1 (z + 2) + A2 .
Setting z = −2 on both sides, we have that

16 − 4 = 12 = A2 .
We can also immediately see from the cross multiplication that

A1 = −8,
by matching the terms on both sides that each multiply z. Putting these terms together, we have the full
partial fraction expansion for Y (z),
−8z 12z
Y (z) = + .
(z + 2) (z + 2)2
Using linearity to invert each term of the z-transform independently, we obtain

y[n] = −8(−2)n u[n] − 6n(−2)n u[n].

We now consider a case where the dierence equation contains an input, or drive term, such that we no
longer have a homogenous dierence equation.
Example
Consider the following linear constant coecient dierence equation.
 n
3 1 1
y[n + 2] − y[n + 1] + y[n] = u[n], y[0] = 4, y[1] = 0.
2 2 3
Taking the unilateral z-transform of both sides and using the advance property, we obtain

 3 1 z
z 2 Y (z) − y[0] − z −1 y[1] − z [Y (z) − y[0]] + Y (z) =

1
2 2 z− 3
3 1 z
z 2 [Y (z) − 4] − z [Y (z) − 4] + Y (z) = 1
2 2 z− 3
 
3 1 z
Y (z) z 2 − z + = 1 + 4z 2 − 6z.
2 2 z− 3

©A.C Singer and D.C. Munson, Jr. March 12, 2011

118 z-transform

We can now solve for Y (z) and keep the terms on the right hand side separated into two distinct groups,
namely,
 

1  z 
Y (z) = + 4z 2 − 6z .
 
z 2 − 32 z + 12  z−1

 | {z }
| {z 3}

term due to initial conditions
term due to input

We can now write the z-transform as a sum of two terms, one due to the input, and one due to the initial
conditions. Recall from our analysis of linear constant coecient dierence equations that these correspond
to the zero-state response and the zero-input response of the system. Taking these two terms separately,
again through linearity of the transform, we have that

Y (z) = Tz (z) + T2 (z)

where

z
T1 (z) = 3

,
+ 12 z − 31


z2 − 2z
4z 2 − 6z
T2 (z) =
.
z − 32 z + 12
2

Here, T1 (z) is the z-transform of the zero-state repsonse, and T2 (z) is the z-transform of the zero-input
response. We can then take a pratial fraction expanson of each of the terms independently. For the rst
term, we nd it convenient to express the partial fraction expansion as

T1 (z) 1 A1 A2 A3
= = + + .
z (z − 12 )(z − 1)(z − 31 ) (z − 21 ) (z − 1) (z − 31 )

This leads to

A1 = −12, A2 = 3, A3 = 9,
−12z 3z 9z
T1 (z) = 1 + (z − 1) + ,
(z − 2 ) (z − 31 )

and the resulting zero state response is given by

 n  n
1 1
yx [n] = −12 +3+9 , n ≥ 0.
2 3

For the zero-input response term, we have that

T2 (z) 4z − 6 B1 B2
= = + ,
z (z − 12 )(z − 1) (z − 21 ) (z − 1)

from which we can quickly solve for the constants, yielding

B1 = 8, B2 = −4,

which gives the partial fraction expansion for the zero-input response as

8z 4z
T2 =
− .
z − 12 (z − 1)

The resulting zero-input response is then given by

 n
1
ys [n] = 8 − 4, n ≥ 0.
2

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.5 Dierence equations and the z-transform 119

Putting the zero-state response and the zero-input response together, we obtain the total response
 n  n
1 1
y[n] = yx [n] + ys [n] = −4 −1+9 , n ≥ 0.
2 3
In general, this method of solution can be applied to linear constant coecient dierence equations of
arbitrary order. Note that while in this particular case, we applied the time advance property of the
unilateral z-transform, when solving dierence equations of the form

y[n] + a1 y[n − 1] + · · · + aN y[n − N ] = x[n], n ≥ 0,

which initial conditions y[−k],k = 1, . . . , N, we can use the Delay Property #2.

5.5.1 General form of solution of linear constant coecient dierence equations

(LCCDE)s
In this section, we will derive the general form of a solution to a linear constant coecient dierence equation.
We will prove that the zero-state response (response to the input, when state is initially zero) is given by a
convolution. Consider the following dierence equation

y[n + K] + a1 y[n + K − 1] + · · · + aK y[n] = x[n], n ≥ 0

together with initial conditions y[k], k = 0, 1, . . . K − 1. Taking the one-sided z-transform of both sides, and
using the Advance Property, we obtain
" K−1
# " K−2
#
X X
K −m K−1 −m
z Y (z) − y[m]z + a1 z Y (z) − y[m]z + · · · + aK−1 z [Y (z) − y[0]] + aK Y (z) = X(z).
m=0 m=0

By dening
"K−1 # "K−2 #
X X
K −m K−1 −m
S(z) = z y[m]z + a1 z y[m]z + · · · + aK−1 zy[0],
m=0 m=0
we have that
Y (z)[z K + a1 z K−1 + · · · + aK ] = X(z) + S(z),
where the characteristic polynomial is given by

z K + a1 z K−1 + · · · + aK .

We now dene the transfer function H(z) of the system described by the LCCDE as

1
H(z) = .
z K + a1 z K−1 + · · · + aK
We then obtain that
 

Y (z) = H(z)  X(z) + S(z) .

 
| {z } |{z}
term do to the input term due to initial conditions

Notice that the decomposition property holds with

ys [n] = Z −1 {H(z)S(z)}
yx [n] = Z −1 {H(z)X(z)} .

Both homogeneity and superposition hold with respect to ys [n] and yx [n] because the z-tranform is linear.
Linear constant coecient dierence equations (LCCDE)s describe linear systems, which have already ex-
plored the time-domain (sequence-domain). It is worthwhile to consider the form of the solution that ys [n]
will take.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

120 z-transform

Consider rst the case when the roots of the characteristic polynomial are distinct. In this case, we have

S(z)H(z) B1 B2 BK
= + + ··· + .
z (z − r1 ) (z − r2 ) (z − rK )

From the denition of S(z), z is a factor in S(z), so there is no need for a z −1 B0 term in the partial fraction
expansion. Multiplying by z, we have

B1 z B2 z BKz
S(z)H(z) = + + ··· + ,
(z − r1 ) (z − r2 ) (z − rK )

from which we can easily recover the sequence

K
X
ys [n] = Bi (ri )n , n ≥ 0,
i=1

which is in the same form as the homogeneous solution that would be obtained from a time-domain solution
of the LCCDE.
We can now observe the form of yx [n]. Since we have that

yx [n] = Z −1 {H(z)X(z)} ,

the partial fraction expansion shows that yx [n] will involve terms in both y[n] and x[n]. We can also rewrite
yx [n] using the convolution property:

n
X
yx [n] = h[m]x[n − m],
m=0

where
 
H(z)
h[m] = Z −1 {H(z)} = Z −1 z
z
  
D0 D1 DK
= Z −1 z + ++
z z − r1 z − rK
( PK n
D0 + i=1 Di (ri ) , n = 0
= PK n
i=1 Di (ri ) , n ≥ 1.

So, we see that yx [n] is given by a convolution of the input with h[n] = Z −1 {H(z)}. Note that the sequence
h[n], n ≥ 0, can be interpreted as the system unit pulse response (u.p.r), or impulse response, assuming zero
initial conditions.
Denition
The unit-pulse seqeunce, or the discrete-time impulse, is given by
(
1, n=0
δ[n] =
0, n 6= 0.

The system response to a unit pulse, or discrete-time impulse, is given by

n
X
y[n] = yx [n]|assuming zero initial conditions = h[m]δ[n − m] = h[n].
m=0

We can explore the use of the impulse response to derive the response to more general signals through
another example.
Example
Consider the following linear system with input x[n] and output y[n] as shown in Figure 5.4 .

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.6 Two-sided z-transform 121

x[n] −→ −→ y[n]
LSI System

Figure 5.4: A linear shift-invariant system.

Suppose that when the input x[n] = δ[n] with zero initial conditions, then the output satises y[n] = an
for n ≥ 0. Again, assuming zero initial conditions (i.e. that the system is initially at rest ), determine y[n]
due to the input x[n] = bn , n ≥ 0.
Solution
Given h[n] = an , n ≥ 0, we know that the output satises y[n] = yx [n], since the initial conditions are all
zero, i.e. the system is initially at rest. We know from the convolution property that

n n  m
X
m n−m
X
na
y[n] = a b =b
m=0 m=0
b

n+1
bn+1 1 − ab
= a , a 6= b
b 1− b
bn+1 − an+1
= , a 6= b.
b−a

Comments
This discussion and these examples lead us to a number of conclusions about the solutions to linear
constant coecient dierence equations. First, we can show (and we will see in the next sections) that the
solution to a linear constant coecient dierence equation will have a essentially the same form when the
input is merely shifted in time. Also, we will see that a similar form is maintained for inputs that are linear
combinations of shifted versions of the input. For example, the response to an input of the form x[n] will
be similar in form to the response to the input x[n] − 2x[n − 1]. We will also see that the solution methods
developed here, as well as the unilateral z-transform, can be modied to accommodate situations when the
input is applied earlier or later than for n = 0. While we discussed situations here that included both
the zero-input response and the zero-state response, in practice we are generally interested in the zero-state
response, or equivalently, we are interested in the response to an input when the system of interest is initially
at rest. The reason for this is that we either have a system where the initial conditions are all zero, or for
a stable system, such that the roots of the characteristic polynomial are all of modulus less than unity,
|ri | < 1, and that after some time, ys [n] has suciently decayed, such that for time scales of interest for
a given application, y[n] ≈ yx [n]. As a result, from this point forward, we will assume that systems under
discussion are initially at rest, and that all initial conditions are set to zero. As a result, the output of a linear
system will be taken as the zero-state response, and we will be interested in the convolution relationship
between the input and the output.

5.6 Two-sided z-transform

When the input to a discrete-time LSI system is of the form z n for all n, i.e. the two-sided sequence that
has non-zero terms for arbitrarily large positive and negative n, the output of the system is simply a scaled
version of the input. This is the eigenfunction property of LSI systems in discrete-time. The eigenfunction
property of continuous-time systems tells us that when the input to a continuous-time LTI system is of the
form est for all t, then the output will be a scaled version of the input. This is easily shown as a consequence
of the convolution integral for LTI systems

ˆ∞
y(t) = h(τ )x(t − τ )dτ,
−∞

©A.C Singer and D.C. Munson, Jr. March 12, 2011

122 z-transform

where h(τ ) is the impulse repsonse of the continuous-time LTI system. Letting the input take the form of a
complex exponential, we have

ˆ∞
y(t) = h(τ )es(t−τ ) dτ
−∞
ˆ∞
= est h(τ )e−sτ dτ
−∞
st
= e H(s),

where H(s) is the Laplace transform of the impulse response, when the integral exists. We call the signals
of the form est eigenfunctions of continuous-time LTI systems, since they satisfy the property that, when
taken as input to an LTI system, they produce an output that is identical except for a (possibly complex)
scale factor. The scale factor H(s) is called the eigenvalue associated with the eigenfunction. Note that
eigenvalue for a given s is the same as the Laplace transform of the impulse response, evaluated at that value
of s. The only signals that have this property, i.e. the only eigenfunctions for LTI systems, are signals of the
form est , for dierent possible values of the complex parameter s. Note that sinusoids are not eigenfunctions
for LTI systems! That means that if a sinusoid is input to an LTI system, the output will not be a simple
scaled version of the input. However, since a sinusoid can be simply constructed as a sum of two such
eigenfunctions, we can easily see what the output will be:

ˆ∞
y(t) = h(τ ) cos(ω(t − τ ))dτ
−∞
ˆ∞
1  jω(t−τ ) 
= h(τ ) e + e−jω(t−τ ) dτ
2
−∞
ˆ∞ ˆ∞
1  jω(t−τ )  1  −jω(t−τ ) 
= h(τ ) e dτ + h(τ ) e dτ
2 2
−∞ −∞
1 jωt
e H(jω) + e−jωt H(−jω) .


=
2

Now, if the impulse response is a purely real-valued function, then its Fourier transform will have complex
conjugate symmetry, such that

1 jωt
e H(jω) + e−jωt H ∗ (jω)


y(t) =
2
1  jωt 
= e |H(jω)|ej∠H(jω) + e−jωt |H(jω)|e−j∠H(jω)
2
1  
= |H(jω)| ejωt ej∠H(jω) + e−jωt e−j∠H(jω)
2
= |H(jω)| cos (ωt + ∠H(jω)) .

While the output is not simply a scaled version of the input, when we decompose the sinusoid into a sum
of two eigenfunctions, we can use linearity of the LTI system to construct the output as a sum of the two
eigenfunction outputs.

Returning to discrete-time LSI systems, when the input to an LSI system is of the form zn for all n, the

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.6 Two-sided z-transform 123

convolution sum yields that

∞
X
y[n] = h[m]z (n−m)
m=−∞
∞
X
= z n
h[m]z −m
m=−∞
= z n H(z),
when the sum converges. Once again, we call signals of the form zn eigenfunctions of discrete time LSI
systems, and the associated eigenvalues, H(z), correspond to the two-sided z-transform of the impulse
response, evaluated at the particular value of z .
We dene the two-sided z-transform of a sequence y[n] as follows

∞
X
Y (z) = y[n]z −n ,
n=−∞

for values of z for which the sum converges. We call the values of z for which the sum converges the region
of convergence of Y (z), or simply the ROCY . Note that as with the unilateral z-transform, the two-sided
(or bilateral) z-transform is again a complex function of a complex variable, meaning that it can take on
complex values and that its argument is itself a complex variable.
For the two-sided transform, we can consider again a few example sequences for which the sequence values
are non-zero for both positive and negative index values.
Example
Consider the following sequence,
(
n n an , n ≥ 0
y[n] = a u[n] + b u[−n − 1] =
bn , n < 0.

Now, using the denition of the z-transform, we have for this sequence,

∞
X
Y (z) = (an u[n] + bn u[−n − 1]) z −n
n=−∞
X∞ ∞
X
= (an u[n]) z −n (bn u[−n − 1]) z −n
n=−∞ n=−∞
∞
X −1
X
= an z −n bn z −n
n=0 n=−∞
∞  −1  n
X a n X b
= +
n=0
z n=−∞
z
−1  n
z X b
= , |z| > |a| +
z−a n=−∞
z
∞  m
z X z
= , |z| > |a| +
z−a m=1
b
z z
= , |z| > |a| + , |z| < |b|,
z−a b−z
where we must combine the two conditions on |z|, to ensure convergence of both of the summations in
the expression. Otherwise, one of the terms in the expression will be invalid, and the resulting algebraic
expression will not be meaningful. Hence, we have

z z
Y (z) = + , |a| < |z| < |b|.
z−a b−z

©A.C Singer and D.C. Munson, Jr. March 12, 2011

124 z-transform

Figure 5.5: Region of convergence of the two-sided z-transform for a two-sided sequence.

Note that the region of convergence, ROCY , in this case is a ring, or annulus, in the complex plane as shown
in Figure 5.5.
In this example,
R− = |a|, R+ = |b|.
If |a| ≥ |b| then ROCY would be the empty set and z-transform would be undened (i.e. is innite) for
all z . The reason that the region of convergence turns out to be a ring in the complex plane comes from
properties of the summations that were assumed to converge in deriving the algebraic expression for the
resulting z-transform. Specically, looking at the denition of the z-transform, we obtain

−1
X ∞
X
Y (z) = y[n]z −1 + y[n]z −1
n=−∞ n=0
| {z } | {z }
converges for zsmall enough, i.e. |z| < R+ converges for zlarge enough, i.e. |z| > R− .

Note that R− is determined by y[n], n ≥ 0 and R+ is determined by y[n], n < 0. If y[n] = 0 for n < 0, then
we have
∞
X
Y (z) = y[n]z −n
n=0

and R+ = ∞, which is essentially a one-sided (unilateral) z-transform. As a result, the region of convergence
corresponds to |z| > R− , as in Figure 5.6. If y[n] = 0 for n > 0, then we have that
0
X
Y (z) = y[n]z −n
n=−∞

and R− = 0, which implies that the region of convergence corresponds to a solid disk in the complex plain,
i.e. we have |z| < R+ as in Figure 5.7. Note that in contrast to the one-sided z-transform, the two-sided
z-transform can accommodate a wider range of signal behaviors, since they can be left-sided, right-sided,
or two-sided and still have a bilateral z-transform. As such, we must state the ROC for Y (z) to uniquely
identify y[n].
A right sided sequence is one that is zero for all n before some time index, i.e. y[n] = 0, n < n0 , for some
n0 . A left-sided sequence is one that is zero for all n after some index, i.e. y[n] = 0, n > n0 , for some n0 , and
a two-sided sequence is one that is neither left-sided nor right sided, i.e. it has non-zero terms for arbitrarily

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.6 Two-sided z-transform 125

Figure 5.6: Region of convergence for a right sided sequence.

Figure 5.7: Region of convergence for a left-sided sequence.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

126 z-transform

large positive and negative indices. Examples of a right-sided sequence, include the unit step sequence, u[n],
and the complex exponential sequence an u[n]. An example left-sided sequence could be u[−n] or an u[−n−1].
−|n|
A two-sided sequence is one such as a , where, for |a| < 1 is a decaying geometric seqeunce for positive
and negative n. Since the two-sided z-transform multiplies the sequence y[n] by z
n and then sums the
resulting modulated sequence for each value of z , in Y (z), then whether a sequence is left-sided, rightsided
or two-sided play an important role in the convergence (and the ROC) of the z-transform. Specically, a
right-sided sequence will have an innite number of terms for large positive n, and, hence, the z-transform can
converge when the magnitude of z is suciently large that z n dominates, making the sequence convergent.
Therefore, right-sided sequences will have a ROC that is the entire z-plane outside of a circle of some radius
(with the possible exception of innity). Similarly, a left-sided sequence can converge when the magnitude of
z is suciently small, such that z n , for large negative n decays suciently rapidly to dominate, making the
series convergent. Therefore, a left-sided sequence will have a ROC for a disc-shaped region in the complex
plane (with the possible exception of zero). A two-sided sequence, having both left-sided and right-sided
elements must balance the eects such that the ROC will result in an annulus (ring) in the complex plane.
Example
Consider the following two sequences,
(
−(an ), n < 0
x[n] = −(an )u[−n − 1] =
0, n≥0
(
an , n ≥ 0
y[n] = an u[n] =
0, n < 0.
For x[n], we have

∞
X
X(z) = x[n]z −n
n=−∞
−1
X
= − an z −n
n=−∞
−1  n
X a
= −
n=−∞
z
∞  k
X z
= −
a
k=1
z
a z
= z , <1
a−1 a
z
= , |z| < |a|.
z−a
Similarly, we have already seen that
z
Y (z) = , |z| > |a|.
z−a
So, we see that the algebraic form of X(z) and Y (z) are identical, but they are not the same functions, since
they are dened on completely dierent regions of the complex plane. The z-transform of a sequence is not
simply dened by the algebraic expression alone, but rather, the combination of the algebraic expression to-
gether with the region of convergence. In order to uniquely specify a sequence from its z-transform, we must
include both the algebraic form as well as the region of the complex plane over which the form is valid. This
leads to the following set of relations.

uniquely dened sequence ⇐⇒ z-transform and region of convergen

n z
a u[n] ⇐⇒ , |z| > |a|
z−a
z
−(an )u[−n − 1] ⇐⇒ , |z| < |a|
z−a

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.6 Two-sided z-transform 127

Figure 5.8: Region of convergence of the two-sided z-transform of u[n].

Poles and Zeros

When sequences correspond to z-transforms that are rational functions (ratios of nite-order polynomials
in z ), we can explore some of the properties of the sequences and their z-transforms by examining the roots of
the numerator and denominator polynomials. These are referred to as the zeros and the poles, respectively,
of a rational z-transform. Specically, for a z-transform given by

B(z)
X(z) = , z ∈ ROCX ,
A(z)

we refer to the values of z such that B(z) = 0, as the zeros of X(z), and the values of z for which A(z) = 0,
as the poles of X(z). That is,

zeros: = {z : B(z) = 0}
poles: = {z : A(z) = 0},

for rational X(z). Rational z-transforms always have ROCs that are bounded by poles. This means that the
ROC is either a disc, an annulus, or the entire plane minus a disc, with the possible exclusion of zero and
innity.
Example
Consider the rational transform
z
Y (z) = , |z| > 1,
z−1
which has a pole at z = 1. This corresponds to the seqeunce x[n] = u[n]. The region of convergence for the
z-transform is given by |z| > 1 as shown in Figure 5.8.
Example
Consider the sequence with rational transform

z
Y (z) = , |z| < 2,
z−2

which has a pole at z = 2. This corresponds to the seqeunce y[n] = −(2n )u[−n−1]. The region of convergence
is now the disk shown in Figure 5.9 .
Example
©A.C Singer and D.C. Munson, Jr. March 12, 2011
128 z-transform

Figure 5.9: Region of convergence of the sequence −(2n )u[−n − 1].

Now consider the sequence with rational transform

z
Y (z) = 2, |z| < 2,
(z − 2)

which has a second-order pole at z = 2. For muiltiple poles and a left-sided seqeunce, we use the same
methods we did for the right-sided case. We can easily show that

−az
nan u[−n − 1] ⇐⇒ 2, |z| < |a|.
(z − a)

Thus, we have that

1
y[n] = − n(2n )u[−n − 1].
2
Example
Now consider the sequence with rational transform given by

z z
Y (z) = + , 1 < |z|| < 2,
z−1 z−2
which has poles at z=1 and z = 2. The region of convergence is therefore an annulus in the complex plane,
and the sequence will turn out to be two sided,

(
1, n≥0
y[n] =
−(2n ), n<0
= u[n] − (2n )u[−n − 1].

The region of convergence is depicted in Figure 5.10.

Example
Consider the seqeunce given by

 n
1
x[n] = , −∞ < n < ∞.
3

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.6 Two-sided z-transform 129

Figure 5.10: Region of convergence for the sequence y[n] = u[n] − (2n )u[−n − 1].

For such a two-sided sequence, does the two-sided z-transform, X(z) exist? Let us examine the z-transform
of the seqeunce from the denition, from which we have

∞  n
X 1
X(z) = z −n
n=−∞
3
−1  n ∞  n
X 1 −n
X 1
= z + z −n .
n=−∞
3 n=0
3

From here, we can see that the rst sum will converge for |z| < 13 , but the second sum will only converge
1
for |z| > . As such, there is no value of z for which both sums will converge. Thus, X(z) does not exist for
3
any z. The z-tranform of this sequence cannot be dened, since the sums do not converge.
Example
Now let us consider a slightly dierent variation on the two-sided above, let

 |n|
1
x[n] = , −∞ < n < ∞.
3

For this sequence, we might have some hope of nding a range of values of z for which the z-transform will
converge, since the sequence remains bounded for all n. In this case we write

 n
1
x[n] = u[n] + 3n u[−n − 1].
3

We can transform the right-sided and left-sided pieces individually, and add the results, by linearity of the
transform, taking into account the regions in the complex plane for which the series will converge. Since each
series has a dierent region of convergence, we need to consider, for the total sequence, only that portion of

©A.C Singer and D.C. Munson, Jr. March 12, 2011

130 z-transform

the complex plane that is common to both the ROC for the right-sided part and the left-sided part. That
is, we need to know for which values of the complex plain will the total z-transform converge. This leads us
to the following transform for the sequence:

z z 1
X(z) = 1 − , < |z| < 3.
z− 3
z−3 3

This transform brings to bear an important property of the region of convergence for a two-sided z-transform,
i.e. the two-sided transform of a two-sided sequence. If the algebraic form for a z-transform is A(z), e.g.
X(z) = A(z), z ∈ ROCX , where

N (z)
A(z) = ,
(z − p1 )(z − p2 ) · · · (z − pN )

then ROCX is generally smaller than the set of z where A(z) alone is well dened. Indeed, A(z) is well
dened at all z except the pole locations z = pi , 1 ≤ i ≤ N, whereas ROCX must be a ring in the complex
plane. It is important to remember that the z-transform of a sequence is not dened solely by an algebraic
expression, but rather by the combination of an algebraic expression and the region of the complex plane
over which the expression is correct. Outside of this region, the algebraic expression is not the z-transform
of the sequence of interest. Some points to remember are that

1. Poles cannot lie in ROCX (because even A(z) is undened at the pole locations).

2. ROCX is generally smaller than the set of z where A(z) is dened.

3. The z-transform, X(z), is given by the pair of A(z) and ROCX .

Another example that will illustrate this point follows.
Example
1 n


Let the sequence x[n] be dened as x[n] = 2 u[n]. The z-transform of the seqeunce can readily be
found to be
z 1
X(z) = 1 , |z| > 2 .
z−2
The algebraic form for X(z) is dened everywhere except at z = 21 , and yet, the z-transform is not dened
1 1
for |z| < 2 . For example, consider when z = 4 , for which we can evaluate the algebraic expression to be

z
1 = −1.
z− 2 z= 14

However, this does not imply that X( 41 ) = −1. Indeed, at z = 41 , X(z) is not dened, since this is not in the
region of convergence of the z-transform, i.e.,

  ∞  n ∞
1 X 1 X
X = u[n]z −n = 2n ,
4 n=−∞
2 n=0
z= 41

which clearly fails to converge.

5.7 Properties of the two-sided z-transform

5.7.1 Linearity
When two sequences x[n] and y[n] have a two-sided z-transforms, X(z) and Y (z), respectively, then the
superposition of these sequences will also have a two-sided z-transform, so long as X(z) and Y (z) are jointly
dened on a non-null subset of the z-plane. Specically, we have

w[n] = ax[n] + by[n] ⇐⇒ W (z) = aX(z) + bY (z), ROCW ⊇ ROCX ∩ ROCY ,

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.7 Properties of the two-sided z-transform 131

Figure 5.11: Region of convergence for the sum of two seqeunces.

that is, the region of convergence is at least as large as the intersection ROCX ∩ ROCY .
Example
Let us consider w[n] = x[n] + y[n] with

z
X(z) = , |z| < 2,
(z + 2)(z + 3)
2
Y (z) = , |z| < 2,
z+2
from which we have that

W (z) = X(z) + Y (z)

z + 2(z + 3)
=
(z + 2)(z + 3)
3(z + 2)
=
(z + 2)(z + 3)
3
= .
z+3
Now, the region of convergence of this expression must be determined. We know two things,

1. The ROC is bounded by poles

2. The ROC contains ROCX ∩ ROCy .

For this example, there is a pole at z = −3. We also have that ROCX ∩ ROCY = {z : |z| < 2}as shown in
Figure 5.11 .
We now can see that the proper region of convergence must be ROCW = {z : |z| < 3}. So, the ROC can
be larger than the intersection if we have pole-zero cancellation on the boundary of intersection, in which
case, the ROC expands outward or inward to be bounded by another pole.

5.7.2 Shifting property

For the two-sided z-transform, the shifting properties are much simpler than their counterparts in the unilat-
eral z-transform, since we do not need to worry about terms shifting in-to or out-of the summation dening
the z-transform. We simply have

x[n] ←→ X(z) ⇐⇒ x[n − k] ←→ z −k X(z)

©A.C Singer and D.C. Munson, Jr. March 12, 2011

132 z-transform

and the region of convergence of the shifted sequence remains unchanged, except for the possible addition
or deletion of z=0 or |z| = ∞.
Example
Consder the seqeunce x[n] = δ[n − 2] for which we have Y (z) = z −2 , |z| > 0. now, if we let y[n] =
x[n + 3] = δ[n + 1], then we have Y (z) = z, |z| < ∞. In this case, we see that z = 0 was added to the region
of convergence and |z| = ∞ was removed from the region of convergence. The proof of the shifting property
follows that for the unilateral z-transform, only simpler. We have

∞
X
X(z) = x[n]z −n
n=−∞
X∞
Y (z) = y[n]z −n
n=−∞
X∞
= x[n − k]z −n
n=−∞
X∞
= x[m]z −(m+k)
m=−∞
∞
X
= z −k x[m]z −m
m=−∞
−k
= z X(z),

where in the fourth line, the change of variable m=n−k was made.

5.7.3 Convolution
The convolution property for the two-sided z-transform follows similary from the unilateral case, for which
we have
∞
X
y[n] = h[m]x[n − m] ⇐⇒ Y (z) = H(z)X(z), ROCY ⊇ ROCX ∩ ROCH ,
m=−∞

so long as there exists a non-null intersection ROCX ∩ ROCH . Just as with linearity, if there is pole-zero
cancellation on a boundary of the intersection, then ROCY expands to the next pole.
Example
Consider the seqeunces x[n] and h[n] for which we have z-transforms X(z) and H(z) and dene Y (z) as
follows

Y (z) = H(z)X(z),

where

1
H(z) = , 1 < |z| < 2,
(z + 1)(z + 2)
z+1
X(z) = , |z| < 2.
z+2

Note that ROCH ∩ ROCX = {z : 1 < |z| < 2}, however we have that ROCY = {z : |z| < 2}.
The convolution formula can be readily shown by taking the z-transform of both sides of the convolution
sum. Since each of the steps in this derivation is reversible, this shows the if and only if nature of the

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.8 The system function and poles and zeros of an LSI system 133

convolution property. Specically, we have

∞
X
y[n] = h[m]x[n − m]
m=−∞
∞ ∞
!
X X
Y (z) = h[m]x[n − m] z −n
n=−∞ m=−∞
∞ ∞
!
X X
= h[m] x[n − m]z −n
m=−∞ n=−∞
∞
!
X
−m
= h[m]z X(z)
m=−∞
= H(z)X(z).

5.8 The system function and poles and zeros of an LSI system
The transfer function of an LSI system with input x[n] and output y[n] is dened for two-sided z-transforms
using
Y (z)
H(z) =
X(z) zero initial conditions.
Indeed, we have seen that H(z) is independent of X(z), and therefore independent of x[n]. For an LSI system,
we can nd H(z) by a number of means. For example, we can

1. Directly compute the z-transform of h[n] using the two-sided z-transform.

2. Compute the quantity H(z) = Y (z)/X(z), for a given pair of input and output sequences x[n] and
y[n].
3. Determine H(z) directly from a block diagram description of the LSI system.

To further examine the last option, we will consider in more detail the methods used for analysis of LSI
systems using a block diagram comprising delay, adder, and gain elements in Section 5.10.

5.9 Inverse two-sided z-transform

When taking an inverse two-sided z-transform, we can, once again, consider the complex contour-integral
that denes its direct inversion, or, more simply, use methods such as partial fraction expansion to reduce
a rational z-transform into a superposition of simpler terms, each of which can be inverted one at a time.
Unlinke the unilateral z-transform, for each term in the partial fraction expansion, we now must consider the
region of convergence of the overall transform and select the appropriate inverse transform sequence whose
ROC would intersect with that of the overall transform to be inverted. To capture this notion graphically,
consider Figure 5.12.
The poles that lie outside the ROC, i.e. those poles located such that |pi | > R+ correspond to terms in
the partial fraction expation for which a left-sided inverse must be selected. The poles that lie inside the
ROC, that is those poles located such that |pk | < R− correspond to terms in the partial fraction expation for
which a right-sided inverse must be selected. These facts can be readily deduced as follows. The poles that
lie inside the inner ring, i.e. those for which |pi | < R− must have a term in the partial fraction expansion
for which the ROC for each pole intersects that of the overall z-transform. Since the poles are insize the
ROC, the only possibility (out of the two choices, |z| < |pi | and |z| > |pi |) that could possibly overlap with
that of the overall ROC, R− < |z| < R+ is |z| > |pi |, which imples that each of these poles, labeled pRHS i
n
correspond to right-sided inverse transforms, of the form pi u[n], assuming that the poles are not repeated
roots. Similarly, the poles that lie outside the outer ring, i.e. those for which |pk | > R+ must have a term
in the partial fraction expansion for which the ROC for each pole intersects that of the overall z-transform.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

134 z-transform

Figure 5.12: A graphical representation of the ROC for a two-sided rational z-transform that includes the
locations of the poles.

Since the poles are outside the ROC, the only possibility (out of the two choices, |z| < |pk | and |z| > |pk |)
that could possibly overlap with that of the overall ROC, R− < |z| < R+ is |z| < |pk |, which imples that
each of these poles, labeled pLHS
k correspond to right-sided
n
inverse transforms, of the form − (pk ) u[−n − 1],
assuming again that the poles are not repeated roots.
Example
Let us consider a two-sided z-transform to invert as an example. Let Y (z) be given such that the algebraic
form is as follows
z
Y (z) = .
(z − 1)(z − 2)
From this information alone, we are unable to compute y[n], since there are three dierent regions of conver-
gence that could be possible for this algebraic expression, and each would lead to a distinct, and dierent,
y[n]. The three possibilities are ROC1 = {z : |z| < 1}, ROC2 = {z : 1 < |z| < 2}, and ROC3 = {z : |z| > 2},
depicted in Figure 5.13.
These three possible ROCs lead to three dierent sequences, since we know that ROC1 yields a left-sided
sequence, y1 [n], ROC2 yields a two-sided sequence, y2 [n], and ROC3 yields a right sided sequence, y3 [n].
From the partial fraction expansion, we have

Y (z) A B
= +
z z−1 z−2
−z z
Y (z) = + .
z−1 z−2

The corresponding three inverse transforms would yield,

y1 [n] = u[−n − 1] − (2n ) u[−n − 1],

y2 [n] = −u[n] − (2n ) u[−n − 1],
y3 [n] = −u[n] + (2n ) u[n].

Example
Let us consider another example, this time with the ROC given. Let

z
Y (z) = , 2 < |z| < 3.
(z − 2)(z − 3)(z − 4)

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.9 Inverse two-sided z-transform 135

(a)

(b) (c)

Figure 5.13: Three possible regions of convergence for the algebraic expression for Y (z). Shown in (a) is
ROC1 corresponding to y1 [n], in (b) is ROC2 for y2 [n] and in (c) is ROC3 for y3 [n].

From the partial fraction expansion, we have

1 1
2z z z
Y (z) = − + 2 ,
z−2 z−3 z−4
| {z } | {z } | {z }
right sided left sided left sided

which yields,

1 n 1
y[n] = (2 )u[n] + 3n u[−n − 1] − (4n ) u[−n − 1].
2 2
Example
For another example, we consider a sequence with complex poles, i.e.

1
X(z) = , |z| < 1.
(z + 1)2
1
= ,
(z + j)(z − j)

for which we have

1 A B C
= + +
z(z + j)(z − j) z (z + j) (z − j)
1 − 12 − 12
= + + /
z (z + j) (z − j)

This yields,
1 1
2z z
X(z) = 1 − − 2 ,
z+j z−j
| {z }
left sided

©A.C Singer and D.C. Munson, Jr. March 12, 2011

136 z-transform

x[n] z -k x[n-k] X(z) z -k z -k X(z)

x[n] b b x[n] X(z) b b X(z)

x[n] x[n]+w[n] X(z) X(z)+W(z)

w[n] W(z)
Figure 5.14: Basic elements of a delay-adder-gain owgraph. To the left, the delay, gain, and adder elements
are shown with their corresponding time-domain representation. To the right, the delay, gain and adder
blocks are indicated with their corresponding z-transform representation.

from which we can onbain

1 1
x[n] = δ[n] + (−j)n u[−n − 1] + (j)n u[−n − 1]
2 2
1  −j(π/2)n j(π/2)n

= δ[n] + e +e u[−n − 1]
2  
π
= δ[n] + cos n u[−n − 1]
π  2
= cos n u[−n].
2

5.10 System Block Diagrams

To explore some of the methods for analyzing LSI system properties together with their implementation in
hardware, we often use a delay-adder-gain model or owgraph model for discrete-time LSI structures. In
Figure

Shown in Figure 5.15 is a common delay-adder-gain block diagram for a second-order LSI system. In
the gure, the notation for a delay element is that of a box labeled with z −1 inside. This is to denote that
the operation of a delay element in the z-transform domain (through the delay property of z-transforms)
is to multiply the input by z −1 . For example, the rst delay element in the owgraph, to the left, takes
as input x[n], which we depict in the z-transform domain as X(z). The output of the delay element is the
signal x[n − 1], i.e. the signal x[n] delayed by one time unit. In the z-transform domain we write x[n − 1] as
z −1 X(z).
The transfer function of the LSI system shown in Figure 5.15 can be shown to be

Y (z) b0 + b1 z −1 + b2 z −2
H(z) = = .
X(z) 1 − a1 z −1 − a2 z −2

This can be shown as follows. First, we note that the owgraph structure has only one adder node. If we
write an equation for the output of the adder node as a function of its inputs, and do so using z-transform

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.10 System Block Diagrams 137

x[n] y[n]
b0

z-1 z-1

b1 a1

z-1 z-1

b2 a2

Figure 5.15: A direct-form I structure is a common delay-adder-gain model. Shown is a second-order DFI
structure.

x[n] b0 y[n]

z-1
a1 b1
z-1

a2 b2

Figure 5.16: A delay-adder-gain model for a second order direct form II structure.

domain representation, using linearity and the delay property, we obtain

Y (z) = b0 X(z) + b1 z −1 X(z) + b2 z −2 X(z) + a1 z −1 Y (z) + a2 z −2 Y (z).

Y (z) 1 − a1 z − a2 z −2 =
−1
X(z) b0 + b1 z −1 + b2 z −2
   

Y (z) b0 + b1 z −1 + b2 z −2
H(z) = = .
X(z) 1 − a1 z −1 − a2 z −2
A second structure, called a direct form II structure is shown in Figure 5.16.
This structure can also be shown to have the same transfer function given by

b0 + b1 z −1 + b2 z −2
H(z) =
1 − a1 z −1 − a2 z −2
through a method similar to that employed for the direct form I structure. Here we introduce a three-step
method that is systematic and guaranteed to determine H(z) for any cycle-free delay adder gain owgraph.
A cycle-free delay adder gain owgraph is one in which all closed cycles contain at least one delay element.
The three steps are as follows.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

138 z-transform

Figure 5.17: Flowgraph of a causal LSI system.

1. Label the output of each adder node in the owgraph with a unique z-transform domain label.

2. Write an equation setting the output of each adder node in the owgraph to the sum of the inputs to
the adder node.

3. Use the resulting equations to remove all labels except for X(z) and Y (z), to obtain a single input-
output relation from which H(z) can be obtained by setting H(z) = Y (z)/X(z).
The three steps are illustrated here for the direct for II structure. First, we note that there are two adder
nodes in the owgraph. The adder node to the left does not have a label, so we introduce a new sequence
q[n] as its output and label this Q(z) in the z-transform domain. For this node, we obtain

Q(z) = X(z) + a1 z −1 Q(z) + a2 z −2 Q(z).

The output of the adder node to the right has already been labeled y[n], so that in the z-transform domain
we obtain
Y (z) = b0 Q(z) + b1 z −1 Q(z) + b2 z −2 Q(z).
Finally, from these two equations, we can eliminate Q(z) as follows

Q(z) 1 − a1 z −1 − a2 z −2 = X(z)
 

X(z)
Q(z) =
1 − a1 z −1 − a2 z −2
which can then be substituted into the expression for Y (z) to yield

b0 + b1 z −1 + b2 z −2 Q(z)
 
Y (z) =
X(z)
Y (z) = b0 + b1 z −1 + b2 z −2
 
1 − a1 z −1 − a2 z −2
−1 −2
Y (z) b0 + b1 z + b2 z
= ,
X(z) 1 − a1 z −1 − a2 z −2
as before. To futher illustrate this method, we consider another example.
Example
Consider the LSI system shown in Figure 5.17 .
The rst step in our three step method is to label the outputs of each of the adder nodes. The rst adder
node to the left has q[n] as its output and the second adder node has y[n] as its output. For the rst adder
node, we have
Q(z) = X(z) + az −1 Q(z) + cz −1 Y (z)
and for the second adder node, we have

Y (z) = Q(z) + bz −1 Y (z).

Solving for Q(z), we have

Q(z) = Y (z) 1 − bz −1 .

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.10 System Block Diagrams 139

Plugging this into the other expression, we have

Y (z) 1 − bz −1 1 − az −1 = X(z) + cz −1 Y (z)

Y (z) 1 − (a + b − c)z −1 + abz −2

= X(z)
Y (z) 1
H(z) = = .
X(z) 1 − (a + b − c)z −1 + abz −2

Note that the impulse response h[n] and the system transfer function H(z) are input-output descriptions of
discrete-time LSI systems. These are also called digital lters. Given an input x[n] we can use either the
impulse response to determine the output y[n] through the convolution sum or we can use the system transfer
function to compute the output through the z-transform. In this sense, both h[n] and H(z) summarize the
behavior of the LSI system. However neither tells use what the internal structure of the digital lter
is. Indeed, for any given system transfer function H(z), there are an unlimited number of possible lter
structures that have this same transfer function. For a second-order transfer function of the form

b0 + b1 z −1 + b2 z −2
H(z) =
1 − a1 z −1 − a2 z −2
just two of the possible realizations are the direct form I and direct form II structures we have just visited.
At this point, you may wonder how the lter structure or delay-adder-gain owgraph relates to the actual
lter implementaiton. The answer to this is multifacted. For example, let us consider the direct form I
structure of Figure 5.15.
If the direct form I structure is implemeted in a digital signal processing microprocessor, then we note
that there is a system clock that guides the operation of the lter. While the clock is not shown in the
owgraph, we know that the operation of the system depends on shifting values of the input into the system
and computing values of the output that are then shifted out. It may take several clock cycles (microprocessor
instructions) to compute each single value of the output sequence y[n]. For example, if the DSP has a single
multiplier/accumulator (MAC), then the clock might trigger the following sequence of instructions

1. multiply x[n] by b0
2. multiply x[n − 1] by b1 and add the result to 1)
3. multiply x[n − 2] by b2 and add the result to 2)
4. multiply y[n − 1] by b1 and add the result to 3)
5. multiply y[n − 2] by b2 and add the result to 4) to give y[n].
The values of x[n], x[n − 1], x[n − 2], y[n − 1], y[n − 2] are each stored in memory locations. You might
expect that after y[n] is computed, then in preparation for computing y[n + 1] we would use a sequence of
instructions to move x[n + 1] into the old location for x[n], move x[n] into the old location for x[n − 1], move
x[n − 1] into the old location for x[n − 2], move y[n] into the old location for y[n − 1], and move y[n − 1] into
the old location for y[n − 2]. However, especially in higher order lters, this would be a huge waste of clock
cycles. Instead, a pointer is used to address the proper memory location at each clock cycle. Therefore, it is
not necessary to move data from memory location to memory location after computer each y[n].
Just as there are a large number of lter structures that implement the same transfer, there are many
algorithms (for a specic DSP) that can implement a given lter structure. Two important factors that you
might consider in selecting a particular algorithm are the speed (number of clock cycles required to compute
each output value) and the errors introduced through nite-precision eects, due to nite length registers
used to represent the real-valued coecients of the lter as well as the sequence values. We have not yet
discussed nite register length eects, i.e. that the DSP has nite length registers for both memory locations
as well as for the computations in the arithmetic units. This means that the digital ltering algorithm is not
implemented in an exact manner. There will be error at the lter output due to coecient quantization, and
arithmetic roundo. Of course, longer register lengths will reduce the error at the lter output. Generally,
there is a tradeo between algorithm speed and numerical precision. For a xed register length, error
usually can be reduced by using a more complicated (than Direct Form I or II) lter structure, requiring

©A.C Singer and D.C. Munson, Jr. March 12, 2011

140 z-transform

Figure 5.18: System owgraph example.

more multiplications, additions, and memory locations. This in turn reduces the speed of the lter. The
lter structure used in practice depends on H(z) (some transfer functions are more dicult to implement
with low error), on the available register length, and on the number of clock cycles available per output.
Example
Find the transfer function of the system in Figure 5.18 and construct a Direct Form II lter structure
that implements the same transfer function.
We immediately label the output of the two adder nodes with the labels y[n] and q[n]. From these we
can then write

Y (z) = 6Q(z) + 4X(z)

2
Q(z) = 2X(z) − 3z −1 Q(z) + z −1 Y (z).
5
We can reduce these equations using

2
Q(z) 1 + 3z −1 2X(z) + z −1 Y (z)


=
5
2X(z) + 52 z −1 Y (z)
Q(z) =
(1 + 3z −1 )

which yields

2X(z) + 52 z −1 Y (z)
Y (z) = 6 + 4X(z)
(1 + 3z −1 )
 12 −1   
5 z 12
Y (z) 1 − = + 4 X(z)
1 + 3z −1 (1 + 3z −1 )
 
12
Y (z) (1+3z )−1 + 4
H(z) = =  12 −1

X(z) 5 z
1 − 1+3z −1

16 + 12z −1


H(z) = .
1 + 35 z −1

The Direct Form II structure having this transfer function is now given in Figure 5.19 .
This structure is far simpler than the previous one and it computes exactly the same output y[n]. It is
important to note that digital lter structures cannot have delay-free loops.
Example
Consider the lter structure shown in Figure 5.20.
This owgraph depicts a system that is unrealizable. If we attempt to determine the input-output
relation, we nd
y[n] = x[n] + 3y[n] + 2y[n − 1],
however the adder node has a delay-free loop which implies that the output at time n requires the addition
of terms that include the output at time n. It is impossible therefore to compute y[n] at any n.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.10 System Block Diagrams 141

Figure 5.19: Direct Form II structure for this example.

Figure 5.20: An unrealizable digital lter structure.

Consider the system shown in Figure 5.21below.

We can immediately write that
W (z) = H1 (z)X(z)
and that
Y (z) = H2 (z)W (z)
which leads to

Y (z) = H2 (z)H1 (z)X(z)

Y (z)
= H(z) = H2 (z)H1 (z) = H1 (z)H2 (z),
X(z)
where the last line follows from commutativity of multiplication of z-tranforms. This is known as a cascade
combination of two LSI systems.
Consider the system shown in Figure 5.22below.
We can immediately write that

Y (z) = H1 (z)X(z) + H2 (z)X(z)

= (H1 (z) + H2 (z)) X(z)

which yields that

Y (z)
H(z) = = (H1 (z) + H2 (z)) X(z).
X(z)
This is known as a parallel combination of two LSI systems.
A feedback connection of two LSI systems is depicted in Figure5.23 .

w[n]
x[n] −→ −→ −→ y[n]
H1 (z) H2 (z)

Figure 5.21: A cascade of two LSI systems.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

142 z-transform

−→ H1 (z) −→
x[n] −→ | ⊕ −→ y[n]
−→ H2 (z) −→

Figure 5.22: A cascade of two LSI systems.

Figure 5.23: A feedback connection of two LSI systems.

The transfer function for a feedback connection of LSI systems can readily be obtained by again labeling
the output of the adder node and writing an equation for its output. In this case, we have

W (z) = X(z) + G(z)Y (z)

and we have that

Y (z) = F (z)W (z)
which leads to

Y (z) = F (z) (X(z) + G(z)Y (z))

Y (z) (1 − F (z)G(z)) = F (z)X(z)
F (z)
Y (z) = X(z)
1 − F (z)G(z)

and nally,
F (z)
H(z) = .
1 − F (z)G(z)
We see that for a feedback connection, the overall transfer function is given by the so-called open loop gain
F (z) divided by one minus the closed loop gain, i.e. 1 − F (z)G(z).

5.11 Flowgraph representations of complex-valued systems

5.12 System analysis
As we have seen, the input-output relationship of a linear-shift invariant (LSI) system is captured through its
response to a single input, that due to a discrete-time impulse, or the impulse response of the system. There
are a number of important properties of LSI systems that we can study by observing properties of its impulse
response directly. Perhaps one of the more important properties of such systems is whether or not they are
stable, that is, whether or not the output of the system will remain bounded for all time when the input to
the system is bounded for all time. While for continuous-time systems and circuits stability may be required
for ensuring that components do not become damaged as voltages or currents grow unbounded in a system,
for discrete-time systems, stability can be equally important. For example, practical implementations of
many discrete-time systems involve digital representations of the signals. To ensure proper implementation
of the operations involved, the numerical values of the signal levels must remain within the limits of the

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.13 BIBO stability 143

Figure 5.24: Bounded input x[n], such that |x[n]| < α.

number system used to represent the signals. If the signals are represented using xed-point arithmetic,
there may be strict bounds on the dynamic range of the signals involved. For example, any real number
−1 ≤ x[n] ≤ 1 can be represented as an innite binary string in two's complement notation as

N
X
x = −b0 + bk 2−k .
k=1

In a practical implementation, only nite-precision representations are available, such that all values might
be represented and computed using xed-point two's compliment arithmetic where any signal at a given
point in time would be represented as a B + 1-bit binary string −1 ≤ x[n] < 1,
B
X
x = −b0 + bk 2−k .
k=1

Now, if the input signal such a system was carefully conditioned such that it was less than 1 in magnitude,
it is important that not only does the output remain less than 1 in magnitude, but also all intermediate
calculations must also. If not, then the numbers would overow, and produce incorrect results, i.e. they
would not represent the true output of the LSI system to the given input. If the discrete-time system
were used to control a mechanical system such as an aircraft, such miscalculations due to instability of the
discrete-time system could produce erratic or even catastrophic results.

5.13 BIBO stability

A system is bounded-input, bounded-output (BIBO) stable if for every bounded input, x[n], the resulting
output, y[n], is bounded. That is, if there exists a xed positive constant α, such that

|x[n]| < α < ∞, for all n,

then there exists a xed positive constant β, such that

|y[n]| < β < ∞, for all n,

where the constants α and β are xed, meaning that they do not depend on n. Graphically, if every bounded
input x[n] as shown in Figure 5.24
causes a bounded output y[n] as shown in Figure 5.25

©A.C Singer and D.C. Munson, Jr. March 12, 2011

144 z-transform

Figure 5.25: Bounded output y[n], such that |y[n]| < β.

then the system is BIBO stable.then system is BIBO stable. Note that BIBO stability is a property of
the system and not the inputs or outputs. While it may be possible to nd specic bounded inputs such
that the outputs remain bounded, a system is only BIBO stable if the output remains stable for all possible
inputs. If there exists even one input for which the output grows unbounded, then the system is not stable
in the BIBO sense.
How do we check if a system is BIBO stable? We cannot possibly try every bounded input and check that
the resulting outputs are bounded. Rather, the input-output relationship must be used to prove that BIBO
stability holds. Similarly, the following theorems can be used to provide simple tests for BIBO stability. It
turns out that we can show that BIBO stability can be determined directly from the impulse respnse of an
LSI system. Specically, an LSI system with impulse response h[n] is BIBO stable if and only if the impulse
response is absolutely summable. That is,

∞
X
LSI system is BIBO stable ⇐⇒ |h[n]| < ∞.
n=−∞

To show both sides of the if and only if relationship, we start with assuming that h[n] is absolutely summable,
and seek to show that the output is bounded (suciency). This can be shown directly from the denition
of an LSI system, i.e. from the convolution sum. We can write

∞
X
y[n] = x[n − m]h[m].
m=−∞

Now, we take the absolute value of both sides and obtain

∞
X
|y[n]| = x[n − m]h[m] ,
m=−∞

which can be upper bounded by

∞
X
|y[n]| ≤ |x[n − m]||h[m]|.
m=−∞

Now we want to see that if |x[n]| < αthat we can nd a suitable β such that |y[n]| < β. We have that

∞
X
|y[n]| ≤ α |h[m]|,
m=−∞

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.13 BIBO stability 145

and since we assumed that

∞
X
|h[n]| = γ < ∞,
n=−∞

we have

|y[n]| ≤ αγ = β < ∞.
To show the other direction of the if and only if relation (necessity), we need to show that when the impulse
response is not absolutely summable, then there exists a sequence
Px[n]
∞
that is bounded, but for which the
output of the system is not bounded. That is, given that the sum m=−∞ |h[m]| diverges, we need to show
that there exists a bounded sequence x[n] that produces an output y[n] such that for some xed n0 the
convolution sum diverges, i.e., y[n0 ] is not bounded. From the convolution sum, we have

∞
X
y[n0 ] = x[m]h[n0 − m].
m=−∞

By selecting the sequence x[n] to be such that x[m] = h∗ [n0 − m]/|h[n0 − m]|, (for real-valued h[n], this
amounts to x[m] = sgn (h[n0 − m]) = ±1), then we have that
∞
X h∗ [n0 − m]h[n0 − m]
y[n0 ] =
m=−∞
|h[n0 − m]|
∞
X |h[n0 − m]|2
=
m=−∞
|h[n0 − m]|
∞
X
= |h[n0 − m]|
m=−∞

and letting k = n0 − m, we obtain that

∞
X
y[n0 ] = |h[k]|,
k=−∞

which diverges, completing the proof.

BIBO stablility of a system can also be directly determined from the transfer function H(z), relating
the z-transform of the input to the z-transform of the output. Specically, we have that for an LSI system
with a rational transfer function, the system is BIBO stable if and only if the region of convergence includes
the unit circle. For causal systems, this means that all of the poles of the system are inside the unit circle.
Specically, we have that

An LSI system with transfer function H(z)is BIBO stable ⇐⇒ROCH ⊆ |z| = 1.

We will show this result specically for causal systems, noting that generalizing the result to left-sided and
two-sided seqeunces is straightforward. First, to prove suciency, assume the region of convergence
P∞ ROCH
includes the unit circle. Next, to illustrate that this implies absolute summability, i.e. n=−∞ |h[n]| < ∞,
we consider the poles of the system function. First, the poles (roots of the denominator polynomial) must
lie inside the unit circle since we have assumed that the region of convergence includes the unit circle, and
for causal systems, i.e. systems for which h[n] = 0 for n < 0, we know ROCH is given by |z| > R for some
R > 0. Since this must include the unit circle, then we have that R < 1 and all of the poles lie inside the
unit circle.
The inverse z-transform, as determined by the partial fraction expansion of the system function H(z)
takes the form
N
X
h[n] = bk (pk )n , n ≥ 0,
k=0

©A.C Singer and D.C. Munson, Jr. March 12, 2011

146 z-transform

assuming there are no repeated roots in the denominator polynomial. Since we have that |pk | < 1 for all of
the poles, we know that

∞
X ∞ X
X N
|h[n]| = |bk ||pk |n u[n]
n=−∞ n=−∞ k=0
N
X ∞
X
= |bk | |pk |n u[n]
k=0 n=−∞
N
X |bk |
= < ∞.
1 − |pk |
k=0

For the case of repeated roots, we would simply have to show that series of the form

∞
X
nL (pk )n
k=0

are convergent. This is readily shown by the ratio test, where we compare the (n + 1)th term to the nth
term in the series. Here we have

(n + 1)L |pk |n+1 (n + 1)L

lim = lim |pk | = |pk | < 1,
n→∞ nL |pk |n n→∞ nL
which implies that these series also all converge, indicating that even for repeated roots, we have that a
causal LSI system whose ROC includes the unit circle will have an absolutely summable impulse response,
and therefore will be BIBO stable.
To show necessity, we assume BIBO stability, and hence absolute summability of the impulse response,
and then, for any point z on the unit circle, we have that

∞
X
|H(z)||z|=1 = h[n]z −n
n=0 |z|=1
∞
X
≤ |h[n]||z|−n
n=0 |z|=1
∞
X
≤ |h[n]||1|−n
n=0
X∞
≤ |h[n]| < ∞,
n=0

which implies that the region of convergence includes the unit circle and completes the proof. This indeed
implies that for a causal LSI system with a rational transfer function (in minimal form), the system is BIBO
stable if and only if all of its poles are inside the unit circle.

5.14 System properties from the system function

Some of the properties we have developed are explored in several examples.
Example
Consider the following LSI system with impulse response h[n], we have that

h[n] = cos(θn)u[n]

which leads to
∞
X ∞
X
|h[n]| = | cos(θn)|,
n=−∞ n=0

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.14 System properties from the system function 147

which diverges. Therefore, the system is not BIBO stable.

Example
Consider the following transfer function for a causal LSI system,

z 2 − 3z + 2
H(z) = ,
z 3 − 2z 2 + 21 z − 1

which after factoring the demoninator, yields,

(z − 1)(z − 2) (z − 1)
H(z) = 2 1 = 2 1.
(z + 2 )(z − 2) z +2

We see that H(z) has poles at z = ± √j2 . The system is therefore causal and has all of its poles inside the unit
circle. Therefore the system is BIBO stable. Note that as done in this example, factors that are common to
the numerator and denominator must be cancelled before applying the stability test.
Example
Consider the following system function of an LSI system,

z
H(z) = , |z| < 100.
z + 100
Note that this is a non-causal system, with a left-sided impulse response. The ROC in this case includes the
unit circle, and therefore the system is BIBO stable.
Example
Consider the following impulse response of an LSI system,


n
4 ,

 0 ≤ n ≤ 106 ,
n
h[n] = n 21 , n > 106

0, n < 0.


Testing for absolute summability of the impulse response, we see that

6
∞ 10 ∞  n
X X X 1
|h[n]| = 4n + n < ∞,
n=0 n=0
2
n=106 +1

and therefore the system is BIBO stable.

We continue exploring the properties of LSI systems through observation of their system functions (that is,
the z-transform of the impulse response), with a focus on the relationship between the region of convergence
of the z-transform and the stability and causality of the system.
Example
Consider the following system function of a stable LSI system,

z
H(z) = 1 ,
(z − 4 )(z − 2)

can it be causal?
Answer: No, it cannot be causal. First, note that although the region of convergence is not explicitly
stated, it is implicitly determined. Noting that the system is stable, we know that the region of convergence
must include the unit circle. Given the pole locations, we know that the region of convergence must be
z : 14 < |z| < 2 implying that the impulse response will have leftsided and right-sided components and that
h[n] must be two-sided, i.e. that H(z) is a two-sided z-transform. Since the impulse response is two-sided,
this implies that the system cannot be causal, i.e. h[n] is non-zero for n<0 and from the convolution sum,

∞
X
y[n] = h[m]x[n − m],
k=−∞

we see that this implies that y[n] depends on values of x[m] for m > n.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

148 z-transform

Figure 5.26: Pole-zero plot for an LSI sys-

tem.

Example
Consider the following discrete time system,

2
y[n] = (x[n]) .

Is this system stable?

Answer: This system is not linear. Therefore, we cannot apply a stability test involving either the
impulse response or transfer function, since the tests discussed so far apply only to LSI systems. Since
this system is not LSI, the convolution sum does not hold, so that the input output relationship does not
satisfy y[n] = x[n] ∗ h[n]or Y (z) = H(z)X(z). Instead, we appeal to the denition of BIBO stability.Since
|x[n]| < αfor all n, then we have that |y[n]| < α2 < ∞ for all n. Therefore, the system is indeed stable, albeit
nonlinear.
Unbounded Outputs
Given an unstable LSI system, how do we nd a bounded input that will cause an unbounded output?
This will be illustrated by example for some causal systems in the following examples.
Example
Consider the following causal LSI system with pole-zero plot shown in Figure 5.26 and with system
function H(z) given by
z
H(z) = , |z| > 2.
z−2
The impulse response is therefore given by h[n] = 2n u[n] and is itself unbounded. Since h[n] grows
without bound, almost any bounded input will cause the output to be unbounded. For example, taking
x[n] = δ[n]would yield y[n] = h[n].
Example
Now consider the following LSI system with system function

z
H(z) = , |z| > 1.
z−1
Although the system is not stable, the impulse response remains bounded, as h[n] = u[n], in this case. Here
we could choose x[n] = u[n] (which is bounded) so that y[n] will be a linear ramp in time. Looking at the
z-transform of the output, this corresponds to forcing Y (z) to have a double pole at z = 1, i.e.

z2
Y (z) = H(z)X(z) = ,
(z − 1)

©A.C Singer and D.C. Munson, Jr. March 12, 2011

5.14 System properties from the system function 149

which for the region of convergence of this output corresponds to a sequence that grows linearly in time.
Example
Here we consider an LSI system with a complex-conjugate pole pair on the unit circle. Let

z 2 − z cos(α)
H(z) = , |z| > 1.
(z − ejα ) (z − e−jα )

The complex conjugate pair of poles on the unit circle corresponds to a sinusoidal oscillating impulse response,

h[n] = cos(αn)u[n].

Thinking of the z-transform of the output, note that choosing x[n] = h[n] will cause Y (z) to have double
poles at z = e±jα , which will in turn cause y[n] to have the form of n times cos(αn), which grows unbounded.
From these examples with causal systems, we see that for systems with poles outside the unit circle, since the
impulse response itself grows unbounded, substantial eort would be required to nd a bounded input that
will not cause an unbounded output. For poles on the unit circle, it is more dicult to nd bounded inputs
that ultimately cause the output to be unbounded. In some elds, such as dynamic systems or control, LSI
systems with poles on the unit circle are called  marginally stable systems. In our terminology, they are
simply unstable systems.

©A.C Singer and D.C. Munson, Jr. March 12, 2011

150 z-transform

©A.C Singer and D.C. Munson, Jr. March 12, 2011

 Signals and Systems Analysis
Question Bank with Answer & Explanation

i
 Contents of Course
Signals and Systems Analysis
System Classification and Properties ,Signal Classification and Properties ,Basic Operations on
Signals ,Elementary Signals ,Properties of Systems ,Discrete Time Signals ,Useful Signals
,Discrete-Time Systems in the Time-Domain ,Periodic and Non-Periodic Signals ,Periodic
Signals – 1 ,The Impulse Function ,BIBO Stability & Systems in the Time Domain ,Continuous
Time Convolution ,LTI Systems – Properties ,Fourier Series ,Fourier Series & Coefficients ,
Fourier Series Properties ,Exponential Fourier Series and Fourier Transforms ,Fourier
Transforms ,Inverse Fourier Transform ,Discrete Fourier Transform ,Sampling ,The Laplace
Transform ,ROC Properties ,Inverse Laplace Transform ,Z-Transform ,Inverse Z-Transform ,
Convolution .

ii
SIGNALS
AND
SYSTEMS
ANALYSIS
Contents
System Classification and Properties ............................................................................................................ 1
Signal Classification and Properties .............................................................................................................. 4
Basic Operations on Signals ........................................................................................................................ 12
Elementary Signals ...................................................................................................................................... 17
Properties of Systems ................................................................................................................................. 23
Discrete Time Signals .................................................................................................................................. 33
Useful Signals .............................................................................................................................................. 36
Discrete-Time Systems in the Time-Domain............................................................................................... 40
Periodic and Non-Periodic Signals .............................................................................................................. 44
Periodic Signals ........................................................................................................................................... 48
The Impulse Function.................................................................................................................................. 54
BIBO Stability & Systems in the Time Domain ............................................................................................ 58
Continuous Time Convolution .................................................................................................................... 60
LTI Systems – Properties ............................................................................................................................. 62
Fourier Series .............................................................................................................................................. 64
Fourier Series & Coefficients....................................................................................................................... 66
Fourier Series Properties ............................................................................................................................ 68
Exponential Fourier Series and Fourier Transforms ................................................................................... 71
Fourier Transforms ..................................................................................................................................... 74
Inverse Fourier Transform .......................................................................................................................... 76
Discrete Fourier Transform ......................................................................................................................... 79
Sampling...................................................................................................................................................... 84
The Laplace Transform ................................................................................................................................ 88
ROC Properties ............................................................................................................................................ 93
Inverse Laplace Transform .......................................................................................................................... 97
Z-Transform............................................................................................................................................... 100
Inverse Z-Transform .................................................................................................................................. 102
Convolution ............................................................................................................................................... 105
Wollo University Kombolcha Institute of Technology

System Classification and Properties

1. The type of systems which are characterized by input and the output quantized at certain levels
are called as
a) analog
b) discrete
c) continuous
d) digital
Answer: b
Explanation: Discrete systems have their input and output values restricted to enter some
quantised/discretized levels.

2. The type of systems which are characterized by input and the output capable of taking any
value in a particular set of values are called as
a) analog
b) discrete
c) digital
d) continuous
Answer: d
Explanation: Continuous systems have a restriction on the basis of the upper bound and lower
bound, but within this set, the input and output can assume any value. Thus, there are infinite
values attainable in this system

3. An example of a discrete set of information/system is

a) the trajectory of the Sun
b) data on a CD
c) universe time scale
d) movement of water through a pipe
Answer: b
Explanation: The rest of the parameters are continuous in nature. Data is stored in the form of
discretized bits on CDs.

4. A system which is linear is said to obey the rules of

a) scaling
b) additivity
c) both scaling and additivity
d) homogeneity
Answer: c
Explanation: A system is said to be additive and scalable in order to be classified as a linear
system.

5. A time invariant system is a system whose output

a) increases with a delay in input
b) decreases with a delay in input
c) remains same with a delay in input

Department of Electrical and Computer Engineering 1

Wollo University Kombolcha Institute of Technology

d) vanishes with a delay in input

Answer: c
Explanation: A time invariant system’s output should be directly related to the time of the output.
There should be no scaling, i.e. y(t) = f(x(t)).

6. Should real time instruments like oscilloscopes be time invariant?

a) Yes
b) Sometimes
c) Never
d) They have no relation with time variance
Answer: a
Explanation: Oscilloscopes should be time invariant, i.e they should work the same way
everyday, and the output should not change with the time at which it is operated.

7. All real time systems concerned with the concept of causality are
a) non causal
b) causal
c) neither causal nor non causal
d) memoryless
Answer: b
Explanation: All real time systems are causal, since they cannot have perception of the future,
and only depend on their memory.

8. A system is said to be defined as non causal, when

a) the output at the present depends on the input at an earlier time
b) the output at the present does not depend on the factor of time at all
c) the output at the present depends on the input at the current time
d) the output at the present depends on the input at a time instant in the future
Answer: d
Explanation: A non causal system’s output is said to depend on the input at a time in the future.

9. When we take up design of systems, ideally how do we define the stability of a system?
a) A system is stable, if a bounded input gives a bounded output, for some values of the input
b) A system is unstable, if a bounded input gives a bounded output, for all values of the input
c) A system is stable, if a bounded input gives a bounded output, for all values of the input
d) A system is unstable, if a bounded input gives a bounded output, for some values of the input
Answer: c
Explanation: For designing a system, it should be kept in mind that the system does not blow out
for a finite input. Thus, every finite input should give a finite output.

10. All causal systems must have the component of

a) memory
b) time invariance
c) stability
d) linearity
Answer: a

Department of Electrical and Computer Engineering 2

Wollo University Kombolcha Institute of Technology

Explanation: Causal systems depend on the functional value at an earlier time, compelling the
system to possess memory.

11. Which one of the following is an example of a bounded signal?

a) et coswt
b) et sinw(-t)
c) e-t coswt
d) et cosw(-t)
Answer: c
Explanation: A bounded signal is the one which satisfies the condition |x(t)|< M < ∞ for all t.
Clearly, the signals et coswt, et sinw(-t) and et cosw(-t) are exponentially growing signals as the
power of the function is positive i.e., the signals will grow beyond infinity. Whereas the signal e-t
coswt is an exponentially decaying signal, hence it will decay to zero and will always be less
than infinity. Therefore, it is bounded.

12. A system produces zero output for one input and same gives the same output for several
other inputs. What is the system called?
a) Non – invertible System
b) Invertible system
c) Non – causal system
d) Causal system
Answer: a
Explanation: A system is said to be invertible if the input fed to the system can be retrieved from
the output of the system. Otherwise the system is non-invertible. Also, if a system gives zero
output for any input and gives the same output for many inputs, then the system is non-invertible.

13. Which among the following is a LTI system?

a) dy(t)/dt+ty(t)=x(t)
b) y(t)=x(t)cosπt
c) y(n)=x(n)+nx(n-1)
d) y(n)=x3 (n+1)
Answer: d
Explanation: A system is said to be linear time invariant (LTI) if the input-output characteristics
do not change with time.
This expression has a coefficient which is a function of time. ∴ the system is time variant.
Output when input is delayed by T, y(t,T)=x(t-T)cosπt
If the output is delayed by T, y(t-T)=x(t-T)cosπ(t-T)
Clearly, both expressions are not equal ∴ The system is time variant.
Output when input is delayed by N, y(n,N)=x(n-N)+nx(n-1-N)
If the output is delayed by N, y(n-N)=x(n-N)+(n-N)x(n-1-N)
Clearly, both expressions are not equal ∴ The system is time variant.
Output when input is delayed by N, y(n,N)=x3 (n+1-N)
If the output is delayed by N, y(n-N)= x3 (n+1-N)
Clearly, both expressions are equal. ∴ The system is time invariant

Department of Electrical and Computer Engineering 3

Wollo University Kombolcha Institute of Technology

14. Amplifiers, motors, filters etc. are examples for which type of system?
a) Distributed parameter systems
b) Unstable systems
c) Discrete time systems
d) Continuous time systems
Answer: d
Explanation: Amplifiers, motors, filters etc. are examples of continuous time systems as these
systems operate on a continuous time input signal and produce a continuous time output signal.
Whereas discrete time systems operate on discrete time signals, distributed parameter systems
have signals which are functions of space as well as time and unstable systems produce
unbounded output from bounded or unbounded input.

15. Which among the following systems are described by partial differential functions?
a) Causal Systems and Dynamic systems
b) Distributed parameter systems and linear systems
c) Distributed parameter systems and Dynamic systems
d) Causal systems and linear systems
Answer: c
Explanation: In distributed parameter systems, signals are functions of space as well as time. In
dynamic systems the output depends on past, present and future values of input, hence, both of
these systems are described by differential functions.

16. Which one of the following systems is causal?

a) y(t)=x(t)+x(t-3)+x(t2)
b) y(n)=x(n+2)
c) y(t)=x(t-1)+x(t-2)
d) y(n)=x(2n2)
Answer: c
Explanation: A causal system is one in which the output depends on the present or past values of
the input, not future. If it depends on future values then it is non-causal. For y(t)=x(t)+x(t-
3)+x(t2), y(n)=x(n+2), and y(n)=x(2n2), the output depends on future values i.e., x (t2), x (n + 2)
and x (2n2) respectively. Whereas in y(t)=x(t-1)+x(t-2), the output y(t) depends on past values
only i.e., x(t – 1) and x(t – 2)

Signal Classification and Properties

17. Which of the following signals are monotonic in nature?
a) 1-exp(-t)
b) 1-exp(sin(t))
c) log(tan(t))
d) cos(t)
Answer: a
Explanation: All of the other functions have a periodic element in them, which means the
function attains the same value after a period of time, which should not occur for a monotonic
function.

Department of Electrical and Computer Engineering 4

Wollo University Kombolcha Institute of Technology

18. What is the period of the following signal, x(t) = sin(18*pi*t + 78 deg)?
a) 1⁄9
b) 2⁄9
c) 1⁄3
d) 4⁄9
Answer: b
Explanation: The signal can be expressed as sin(wt + d), where the time period = 2*pi/w.

19. Which of the following signals is monotonic?

a) x(t) = t3 – 2t
b) x(t) = sin(t)
c) x(t) = sin22(t) + cos22(t) – 2t
d) x(t) = log(cos(t))
Answer: c
Explanation: c) reduces to 1 – 2t, which is a strictly decreasing function.

20. For the signal, x(t) = log(cos(a*pi*t+d)) for a = 50 Hz, what is the time period of the signal,
if periodic?
a) 0.16s
b) 0.08s
c) 0.12s
d) 0.04s
Answer: d
Explanation: Time period = 2*pi/(50)pi = 1/25 = 0.04s

21. What are the steady state values of the signals, 1-exp(-t), and 1-k*exp(-k*t)?
a) 1, k
b) 1, 1/k
c) k, k
d) 1, 1
Answer: d
Explanation: Consider limit at t tending to infinity, we obtain 1 for both cases.

22. For a bounded function, is the integral of the function from -infinity to +infinity defined and
finite?
a) Yes
b) Never
c) Not always
d) None of the mentioned
Answer: c
Explanation: If the bounded function, is say y = 2, then the integral ceases to hold. Similarly, if it
is just the block square function, it is finite. Hence, it depends upon the spread of the signal on
either side. If the spread is finite, the integral will be finite.

23. For the signal x(t) = a – b*exp(-ct), what is the steady state value, and the initial value?
a) c, b

Department of Electrical and Computer Engineering 5

Wollo University Kombolcha Institute of Technology

b) c, c-a
c) a, a-b
d) b, a-b
Answer: c
Explanation: Put the limits as t tends to infinity and as t tends to zero.

24. For a double sided function, which is odd, what will be the integral of the function from -
infinity to +infinity equal to?
a) Non-zero Finite
b) Zero
c) Infinite
d) None of the mentioned
Answer: b
Explanation: For an odd function, f(-x) = -f(x), thus the integrals will cancel each other, giving
zero.

25. Find where the signal x(t) = 1/(t2 – 3t + 2) finds its maximum value between (1.25, 1.75):
a) 1.40
b) 1.45
c) 1.55
d) 1.50
Answer: d
Explanation: Differentiate the function for an optima, put it to zero, we will obtain t = 1.5 as the
required instant.

26. Is the signal x(t) = exp(-t)*sin(t) periodic in nature?

a) Yes
b) No
Answer: b
Explanation: Though sin(t) is a periodic function, exp(-t) is not a periodic function, thus leading
to non-periodicity.

27. A signal is a physical quantity which does not vary with ____________
a) Time
b) Space
c) Independent Variables
d) Dependent Variables
Answer: d
Explanation: A signal is a physical quantity which varies with time, space or any other
independent variables. Therefore, it does not vary with dependent variables.

28. Most of the signals found in nature are _________

a) Continuous-time and discrete-time
b) Continuous-time and digital
c) Digital and Analog
d) Analog and Continuous-time

Department of Electrical and Computer Engineering 6

Wollo University Kombolcha Institute of Technology

Answer: d
Explanation: Signals naturally are continuous-time signals. These are also known as analog
signals. Continuous-time or analog signals are defined for all values of time t.

29. Which one of the following is not a characteristic of a deterministic signal?

a) Exhibits no uncertainty
b) Instantaneous value can be accurately predicted
c) Exhibits uncertainty
d) Can be represented by a mathematical equation
Answer: c
Explanation: Deterministic signal is one which exhibits no uncertainty and its instantaneous
value can be accurately predicted from its mathematical equation. Therefore, a deterministic
signal doesn’t exhibit uncertainty. However, a random is always uncertain.

30. Determine the fundamental period of the following signal:sin60t.

a) 1/60 sec
b) 1/30 sec
c) 1/20 sec
d) 1/10 sec
Answer: b
Explanation: Consider the equation: sinΩ0t. Comparing this equation with the one given in the
question: sin60t
⇒ Ω0=60π

31. Sum of two periodic signals is a periodic signal when the ratio of their time periods is
____________
a) A rational number
b) An irrational number
c) A complex number
d) An integer
Answer: a
Explanation: Sum of two periodic signals is a periodic signal only when the ratio of their time
periods is a rational number or it is the ratio of two integers. For e.g., T1/T2 = 5/7 → Periodic;
T1/T2 = 5 → Aperiodic.

32. Determine the Time period of: x(t)=3 cos⁡(20t+5)+sin⁡(8t-3).

a) 1/10 sec
b) 1/20 sec
c) 2/5 sec
d 2/4 sec
Answer: c

Department of Electrical and Computer Engineering 7

Wollo University Kombolcha Institute of Technology

Explanation: Here is the explanation.

33. What is the even component of a discrete-time signal?

Answer: b
Explanation: Here is the explanation.

34. Determine the odd component of the signal: x(t)=cost+sint.

a) sint
b) 2sint
c) cost
d) 2cost
Answer: c
Explanation: Here is the explanation.

35. Is the signal sin(t) anti-symmetric?

a) YES
b) NO
Department of Electrical and Computer Engineering 8
Wollo University Kombolcha Institute of Technology

Answer: a
Explanation: A signal is said to be anti-symmetric or odd signal when it satisfies the following
condition:
⇒ x(t) = – x(t)
Now, here, x(t) = sin(t) ⇒ x(-t) = sin(-t) = – sin(t)
∴ Sin(t) is an anti-symmetric signal or an odd signal.

36. For an energy signal __________

a) E=0
b) P= ∞
c) E= ∞
d) P=0
Answer: d
Explanation: A signal is called an energy signal if the energy satisfies 0<E< ∞ and power P=0.

37. Determine the power of the signal: x(t) = cos(t).

a) 1/2
b) 1
c) 3/2
d) 2
Answer: a
Explanation: Here is the explanation.

38. Is the following signal an energy signal?

x(t) = u(t) – u(t – 1)
a) YES
b) NO
Answer: a

Department of Electrical and Computer Engineering 9

Wollo University Kombolcha Institute of Technology

Explanation: Here is the explanation.

39. What is single-valued function?

a) Single value for all instants of time
b) Unique value for every instant of time
c) A single pattern is followed by after ‘t’ intervals
d) Different pattern of values is followed by after ‘t’ intervals of time
Answer: b
Explanation: Single-valued function means “for every instant of time there exists unique value of
the function”.

40. In real valued function and complex valued function, time is _______________
a) Real
b) Complex
c) Imaginary
d) Not predictable
Answer: a
Explanation: Time is an independent variable and it is real valued irrespective of real valued or
complex valued function. And time is always real.

41. Discrete time signal is derived from continuous time signal by _____________ process.
a) Addition
b) Multiplying
c) Sampling
d) Addition and multiplication
Answer: c
Explanation: Sampling is a process wherein continuous time signal is converted to its equivalent
discrete time signal. It is given by t = N*t.

42. Even signals are symmetric about the vertical axis.

a) True
b) False
Answer: a
Explanation: Signals are classified as even if it has symmetry about its vertical axis. It is given
by the equation x (-t) = x (t).

43. If x (-t) = -x (t) then the signal is said to be _____________

a) Even signal
b) Odd signal
Department of Electrical and Computer Engineering 10
Wollo University Kombolcha Institute of Technology

c) Periodic signal
d) Non periodic signal
Answer: a
Explanation: Signals is said to be odd if it is anti- symmetry over the time origin. And it is given
by the equation x (-t) = -x (t).

44. Which of the following is true for complex-valued function?

a) X (-t) = x*(t)
b) X (-t) = x(t)
c) X (-t) = – x(t)
d) X (-t) = x*(-t)
Answer: a
Explanation: Complex-valued function is said to be conjugate symmetry if its real part is even
and imaginary part is odd and it is shown by the equation x(-t) = x*(t).

45. When x(t ) is said to be non periodic signal?

a) If the equation x (t) = x (t + T) is satisfied for all values of T
b) If the equation x (t) = x (t + T) is satisfied for only one value of T
c) If the equation x (t) = x (t + T) is satisfied for no values of T
d) If the equation x (t) = x (t + T) is satisfied for only odd values of T
Answer: c
Explanation: A signal x (t) is said to be non periodic signal if it does not satisfy the equation x(t)
= x(t + T). And it is periodic if it satisfies the equation for all values of T = T0, 2T0, 3T0…

46. Fundamental frequency x[n] is given by ___________

a) Omega = 2*pi /N
b) Omega = 2*pi*N
c) Omega = 4*pi *2N
d) Omega = pi / N
Answer: a
Explanation: Fundamental frequency is the smallest value of N which satisfies the equation
Omega = 2*pi/ N, Where N is a positive integer.

47. Noise generated by an amplifier of radio is an example for?

a) Discrete signal
b) Deterministic signal
c) Random signal
d) Periodic signal
Answer: c
Explanation: Random signal is the one which there is uncertainty before its actual occurrence.
Noise is a best example for random signal.

48. Energy signal has zero average power and power signal has zero energy.
a) True
b) False
Answer: b

Department of Electrical and Computer Engineering 11

Wollo University Kombolcha Institute of Technology

Explanation: Energy and power signals are mutually exclusive. Energy signal has zero average
power and power signal has infinite energy.

49. What is the fundamental frequency of discrete –time wave shown in fig a?
a) π/6
b) π/3
c) 2π/8
d) π
Answer: b
Explanation: Omega = 2* π / N. In the given example the number of samples in one period is N
= 6. By substituting the value of N =6 in the above equation then we get fundamental frequency
as π/3

Basic Operations on Signals

50. Which of the following is an example of amplitude scaling?
a) Electronic amplifier
b) Electronic attenuator
c) Both amplifier and attenuator
d) Adder
Answer: c
Explanation: Amplitude scaling refers to multiplication of a constant with the given signal.
It is given by y (t) = a x (t). It can be both increase in amplitude or decrease in amplitude.

51. Resistor performs amplitude scaling when x (t) is voltage, a is resistance and y (t) is output
current.
a) True
b) False
Answer: b
Explanation: The given statement is not true. The relation between voltage, current and
resistance is given by V = IR. Comparing with equation y (t) = a x (t), we can see that y (t) is the
output voltage for given current x (t) with resistance R.

52. Which of the following is an example of physical device which adds the signals?
a) Radio

Department of Electrical and Computer Engineering 12

Wollo University Kombolcha Institute of Technology

b) Audio mixer
c) Frequency divider
d) Subtractor
Answer: b
Explanation: Audio mixer is a device which combines music and voice signals. It is given by
Y (t) = x1 (t) + x2 (t).

53. AM radio signal is an example for __________

a) y (t) = a x (t)
b) y (t) = x1 (t) + x2 (t)
c) y (t) = x1 (t) * x2 (t)
d) y (t) = -x(t)
Answer: c
Explanation: AM radio signal is an example for y (t) = x1 (t) * x2 (t) where, x1 (t) consists of an
audio signal plus a dc component and x2 (t) is a sinusoidal signal called carrier wave.

54. Which of the passive component performs differentiation operation?

a) Resistor
b) Capacitor
c) Inductor
d) Amplifier
Answer: c
Explanation: Inductor performs differentiation. It is given by y (t) = L d/dt i(t) where, I (t)
denotes current flowing through an inductor of inductance L.

55. Which of the component performs integration operation?

a) Resistor
d) Diode
c) Capacitor
d) Inductor
Answer: c
Explanation: Capacitor performs integration. V (t) developed across capacitor is given by
v (t) = (1/C)* ∫t-∞ i (∂).d∂, I (t) is the current flowing through a capacitor of capacitance C.

56. Time scaling is an operation performed on _______

a) Dependent variable
b) Independent variable
c) Both dependent and independent variable
d) Neither dependent nor independent variable
Answer: b
Explanation: Time scaling is an example for operations performed on independent variable time.
It is given by y (t) = x (at).

57. Y (t) = x (2t) is ________

a) Compressed signal
b) Expanded signal

Department of Electrical and Computer Engineering 13

Wollo University Kombolcha Institute of Technology

c) Shifted signal
d) Amplitude scaled signal by a factor of 2
Answer: a
Explanation: By comparing the given equation with y (t) = x (at) we get a=2. If a>1 then it is
compressed version of x (t).

58. Y (t) = x (t/5) is _______

a) Compressed signal
b) Expanded signal
c) Time shifted signal
d) Amplitude scaled signal by factor 1/5
Answer: b
Explanation: y (t) = x (at), comparing this with the given expression we get a = 1/5. If 0<a<1
then it is expanded (stretched) version of x (t).

59. In discrete signal, if y [n] = x [k*n] and k>1 then ______

a) Some samples are lost from x [n]
b) Some samples are added to x [n]
c) It has no effect on samples
d) Samples will be increased with factor k
Answer: a
Explanation: For discrete time signal y [n] = x [k*n] and k>1, it will be compressed signal and
some samples will be lost. The samples lost will not violate the rules of sampling theorem

60. In the following diagram, X [n] and y [n] are related by ______

a) Y [n] = 2*x [n]

b) Y [n] = -2*x [n]
c) Y [n] = x [2n]
d) Y [n] = x [-2n]
Answer: a

Department of Electrical and Computer Engineering 14

Wollo University Kombolcha Institute of Technology

Explanation: Y [n] = 2*x [n] is an example for amplitude scaling of discrete time signal. The
given figure is an example for 2*x [n] hence Y [n] = 2*x [n] is correct.

61. X [n] and y [n] is as shown below, the relationship between x [n] and y [n] is given by
______

a) X [n] = y [n]/3
b) X [n] = 3* y [n]
c) Y [n] = x [n]/3
d) Y [n] = 3*x [n]
Answer: c
Explanation: The given y [n] is amplitude scaling of a discrete time signal by a factor 1/3.
Hence the amplitude is reduced by 1/3.

62. Considering figure 3 above, is the following figure true for y [n] = x [2n]?

Department of Electrical and Computer Engineering 15

Wollo University Kombolcha Institute of Technology

a) True
b) False
Answer: a
Explanation: X [2n] is an example of time scaling. For discrete time signal x [k*n], k>1 the
samples will be lost.

63. Considering figure 3 above, is the following figure true for y [n] = x [n/2]?

a) True
b) False
Answer: b
Explanation: X [n/2] is an example for time scaling by factor ½ and it will be a stretched signal.
The discrete time signal should extend from -10 to 10.

64. Consider figure 4, is the given y (t) an integration of x (t)?

a) Y (t) = ∫x (t).dt
b) Y (t) = ∫x2 (t).dt
c) Y (t) = 3* ∫x (t).dt
d) Y (t) = 3* ∫x2 (t).dt
Answer: a
Explanation: The given y (t) is integral of x (t) and amplitude 3 remains constant for t>1.
It is because of the properties of integration.

Department of Electrical and Computer Engineering 16

Wollo University Kombolcha Institute of Technology

65 . Consider figure 4, is the given y (t) a differentiation of x (t)?

Answer: a
Explanation: The given y (t) is differentiation of x (t) and hence we have impulses at -1, 0 and 1.

66. The given pair x (t) and y (t) is _______

a) Y (t) = d/dt (x (t))

b) Y (t) = ∫x (t).dt
c) Y (t) = x (t) -1
d) Y (t) = x (t) /2
Answer: a
Explanation: The given pair x (t) and y (t) is related by y (t) = d/dt (x (t)). From -2 to 2 we have
Y (t) is zero because differentiation of constant is zero.

Elementary Signals
67. The general form of real exponential signal is________
a) X (t) = beat
b) X (t) = (b+1)eat
c) X (t) = b (at)
d) X (t) = be (a+1)t
Department of Electrical and Computer Engineering 17
Wollo University Kombolcha Institute of Technology

Answer: a
Explanation: X (t) = beat is the most general way of representing the exponential signals where
both b and a are real parameters.

68. In the equation x (t) = beat if a < 0, then it is called ______

a) Growing exponential
b) Decaying exponential
c) Complex exponential
d) Both Growing and Decaying exponential
Answer: b
Explanation: If a > 0 in x (t) = beat it is called growing exponential and if <0 it is called decaying
exponential. Hence Decaying exponential is correct.

69. In the below figure if R value is increased then which of the following is true?

a) Slower the rate of decay of v (t)

b) Greater the rate of decay of v (t)
c) Decay rate is independent of R
d) Decay rate depends only on the capacitor value
Answer: a
Explanation: In the circuit shown voltage across capacitor decays exponentially with time at a
rate determined by time constant RC. Hence the larger the resistor, the slower will be the rate of
decay of v (t) with time.

70. The time period of continuous-time sinusoidal signal is given by _____

a) T = 2π / w
b) T = 2π / 3w
c) T = π / w
d) T = π / 2w
Answer: a
Explanation: X (t) = A cos (wt+φ) is the continuous-time sinusoidal signal and its period is given
by
T = 2π / w where w is the frequency in radians per second.

71. The natural angular frequency of the parallel LC circuit is?

Department of Electrical and Computer Engineering 18

Wollo University Kombolcha Institute of Technology

Answer: a
Explanation: Wo is the natural angular frequency and for parallel LC circuit it is given by
wo=1⁄√LC where, L is value of inductor and C is value of capacitor.

72. X [n] = 2 cos (2n) is periodic or not?

a) Periodic with period 2n
b) Periodic with period 2π
c) Periodic with period 2
d) Non periodic
Answer: d
Explanation: The given signal x [n] is non periodic as it doesn’t satisfy the equation w=2πm⁄N
where, N is fundamental period and m is an integer.

73. Check whether x [n] = 7 sin (6πn) is periodic and if it is period calculate its fundamental
period?
a) Periodic with fundamental period 6π
b) Periodic with fundamental period 3
c) Periodic with fundamental period 1
d) Non periodic
Answer: c
Explanation: X [n] = 7 sin (6πn) is a periodic discrete time signal with period 1. By substituting
w = 6π and m=3 in w=2πm⁄N we get N =1.

74. Find the smallest angular frequency for which the discrete time signal with fundamental
period N=8 would be periodic?
a) π⁄4
b) π⁄2
c) 3π⁄4
d) π⁄16
Answer: a
Explanation: By substituting N=8 and m=1 in the equation w=2πm⁄N we get the smallest angular
frequency as π⁄4.

75. Euler’s identity ejθ is expanded as _____

a) cos θ + j sin θ
b) cos θ – j sin θ
c) cos θ + j sin 2θ
d) cos⁡ 2θ+j sinθ
Answer: a

Department of Electrical and Computer Engineering 19

Wollo University Kombolcha Institute of Technology

Explanation: The complex exponential ejθ is expanded as cos θ + j sin θ and is called Euler’s
identity with cos θ as real part sin θ as imaginary part.

76. Exponentially damped sinusoidal signal is ______

a) Periodic
b) Non periodic
c) Insufficient information
d) Maybe periodic
Answer: b
Explanation: Exponentially damped sinusoidal signal of any kind is not periodic as it does not
satisfy the periodicity condition.

77. Mathematical representation of given rectangular pulse is ______

a) X (t) = {2A, t/2 < 0 < -t/2

b) X (t) = {2A, -t/2 < 0 < t/2
c) X (t) = {2A, 0 <= |t| <= t/2
{0, |t| > t/2
d) X (t) = {2A, 0 <|t| < t/2
{0, |t| > t/2
Answer: c
Explanation: The given rectangular pulse is of amplitude 2A for the time interval –t/2 to t/2 and
zero otherwise.

78. If describe x [n] as superposition of two step functions.

a) X [n] = u [n] – u [n-5].
b) X [n] = u [n] + u [n-5].
c) X [n] = u [n-5] – u [n].
d) X [n] = u [n-5] + u [n].
Answer: a
Explanation: X [n] will be of amplitude for the interval 0 to 4 and zero otherwise. It can be
obtained by the equation x [n] = u [n] – u [n-5].

Department of Electrical and Computer Engineering 20

Wollo University Kombolcha Institute of Technology

79 Discrete-time version of unit impulse is defined as ______

Answer: a
Explanation: Unit impulse is an elementary signal with zero amplitude everywhere except at n =
0.

80. Which of the following is not true about unit impulse function?

Answer: d
Explanation: One option gives the definition of discrete-time version of impulse function, other
options gives continuous-time representation of impulse function.

81. The step function u (t) is integral of _______ with respect to time t.
a) Ramp function
b) Impulse function
c) Sinusoidal function
d) Exponential function
Answer: b
Explanation: Step function is an integral of impulse function and conversely, impulse is the
derivative of step function u (t).

82. The area under the pulse defines _____ of the impulse.
a) Strength
b) Energy
c) Power
d) Duration
Answer: a
Explanation: The area under the pulse defines strength of the impulse and the strength of the
impulse is denoted by the label next to the arrow.

Department of Electrical and Computer Engineering 21

Wollo University Kombolcha Institute of Technology

83. Unit impulse ∂(t) is _____ of time t.

a) Odd function
b) Even function
c) Neither even nor odd function
d) Odd function of even amplitude
Answer: b
Explanation: For an impulse function, ∂(-t)= ∂(t). Hence unit impulse is an even function of time
t.

84. Shifting property of impulse ∂(t) is given by ______

Answer: a
Explanation: X (t) be a function and the product of x (t) with time shifted delta function ∂(t – to)
gives x(to), this is referred to as shifting property of impulse function.

85. ∂(at) = 1⁄a ∂(t), this property of unit impulse is called ______
a) Time shifting property
b) Time scaling property
c) Amplitude scaling property
d) Time reversal property
Answer: b
Explanation: Impulse function exhibits shifting property, time scaling property. And time scaling
property is given by∂(at) = 1⁄a ∂(t).

86. Which of the following is not true about the ramp function?

a)
b) r (t) = t u (t)
c) Ramp function with unit slope is integral of unit step
d) Integral of unit step is a ramp function of unit slope
Answer: d
Explanation: The impulse function is derivative of the step function. In the same way the integral
of step function is a ramp function of unit slope.
∫u(t) = r(t).

Department of Electrical and Computer Engineering 22

Wollo University Kombolcha Institute of Technology

Properties of Systems
87. Is the system y(t) = Rx(t), where R is a arbitrary constant, a memoryless system?
a) Yes
b) No
Answer: a
Explanation: The output of the system depends on the input of the system at the same time
instant. Hence, the system has to be memoryless.

88. Does the following discrete system have the parameter of memory, y[n] = x[n-1] + x[n] ?
a) Yes
b) No
Answer: a
Explanation: y[n] depends upon x[n-1], i.e at the earlier time instant, thus forcing the system to
have memory.

89. y[t]= ∫x[t],t ranges from 0 to t. Is the system a memoryless one?

a) Yes
b) No
c) Both memoryless and having memory
d) None of the Mentioned
Answer: b
Explanation: While evaluating the integral, it becomes imperative to know the values of x[t]
from 0 to t, thus making the system requiring memory.

90. y(t) = sin(x(t-1)) : Comment on its memory aspects.

a) Having memory
b) Needn’t have memory
c) Memoryless system
d) Time invariant system
Answer: a
Explanation: The output at any time t = A, requires knowing the input at an earlier time, t = A –
1, hence making the system require memory aspects.

91. Construct the inverse system of y(t) = 2x(t)

a) y(t) = 0.5x(t)
b) y(t) = 2x(t)
c) y(2t) = x(t)
d) y(t) = x(2t)
Answer: a
Explanation: Now, y(t) = 2x(t) => x(t) = 0.5*y(t)
Thus, reversing x(t) <-> y(t), we obtain the inverse system: y(t) = 0.5x(t)

92. y(t) = x2(t). Is y(t) = sqrt(x(t)) the inverse of the first system?
a) Yes

Department of Electrical and Computer Engineering 23

Wollo University Kombolcha Institute of Technology

b) No
c) Inverse doesn’t exist
d) Inverse exist
Answer: b
Explanation: We cannot determine the sign of the input from the second function, thus, the
output doesn’t replicate the input. Thus, the second function is not an inverse of the first one.

93. Comment on the causality of y[n] = x[-n].

a) Time invariant
b) Causal
c) Non causal
d) Time varying
Answer: c
Explanation: For positive time, the system may seem to be causal. However, for negative time,
the output depends on time at a positive sign, thus being in the future, enforcing non causality.

94. y(t) = x(t-2) + x(2-t). Comment on its causality:

a) Causal
b) Time variant
c) Non causal
d) All of the mentioned
Answer: c
Explanation: For a time instant existing between 0 and 1, it would depend on the input at a time
in the future as well, hence being non causal.

95. Comment on the causality of y[n] = n*x[n].

a) Time invariant
b) Time varying
c) Non causal
d) Causal
Answer: d
Explanation: For positive time, the system may seem to be causal. For negative time, the output
depends on the same time instant, thus making it causal.

96.. Comment on the linearity of y[n] = n*x[n].

a) Linear
b) Only additive
c) Not scalable
d) Non linear
Answer: d
Explanation: The function obeys the scaling/homogeneity property, but doesn’t obey the
additivity property, thus not being linear.

97. What is the following type of system called? y[n] = x[n] + y[n-1].
a) Subtractor system
b) Adder system

Department of Electrical and Computer Engineering 24

Wollo University Kombolcha Institute of Technology

c) Product System
d) Divisor System
Answer: b
Explanation: If we write for n-1, n-2, .. we will obtain y[n] = x[n] + x[n-1] + x[n-2] …,
thus obtaining an adder system.

98. Which of the following systems is linear?

a) y(t) = sin(x(t))
b) y(t) = log(x(t))
c) y(t) = cos(x(t))
d) y(t) = dx(t)/dt
Answer: d
Explanation: Only d satisfies both the scaling and the additivity properties.

99. Which of the following systems is stable?

a) y(t) = log(x(t))
b) y(t) = exp(x(t))
c) y(t) = sin(x(t))
d) y(t) = tx(t) + 1
Answer: c
Explanation: Stability implies that a bounded input should give a bounded output. In a, b, d there
are regions of x, for which y reaches infinity/negative infinity. Thus the sin function always stays
between -1 and 1, and is hence stable.

100. Which of the following systems is time invariant?

a) y(t) = x(2t) + x(t)
b) y(t) = x(t) + x(1-t)
c) y(t) = -x(t) + x(1-t)
d) y(t) = x(t) + x(t-1)
Answer: d
Explanation: In each of a, b and c there is a negative sign of t involved, which means a backward
shift of t-0 in time, would mean a forward shift in each of them. However, only in d, the
backward shift will remain as backward, and undiminished.

101. State whether the differentiator system is causal or not.

a) True
b) False
Answer: b
Explanation: The derivative of a function can be written in forward difference and in backward
difference form, hence the derivative would depend on a slightly forward value of the function,
thus making it non causal.

102. State whether the differentiator system is a stable system or not.

a) True
b) False
Answer: b

Department of Electrical and Computer Engineering 25

Wollo University Kombolcha Institute of Technology

Explanation: The derivative of a function can be unbounded at some bounded inputs, like tan(x)
at x=pi/2, hence the differentiator system is unstable in general, when the input is not specified.

103. Which of the following systems is memoryless?

a) y(t) = x(2t) + x(t)
b) y(t) = x(t) + 2x(t)
c) y(t) = -x(t) + x(1-t)
d) y(t) = x(t) + 2x(t+2)
Answer: b
Explanation: A system possessing no memory has its output depending upon the input at the
same time instant, which is prevalent only in option b.

104. For what value of k, will the following system be time invariant?
y(t) = x(t) + x(kt) – x(2t) + x(t-1)
a) 1
b) 2
c) 3
d) 2.5
Answer: b
Explanation: A system possessing no memory has its output depending upon the input at the
same time instant, which is prevalent only in option b.

105. State if the following system is periodic or not. y(t) = sin(sqrt(2)*x(t))

a) No
b) Yes
Answer: a
Explanation: The function y = sin(nx) is periodic only for rational ‘n’.

106. State whether the following system is periodic or not. y(t) = log(sin(x(t)).
a) Yes
b) No
Answer: b
Explanation: Sin x is a periodic function, but log x is not a periodic function. Thus y is log t,
where t= sin x, thus y is not periodic.

107. Which one of the following is an example of a system with memory?

a) Identity System
b) Resistor
c) y(n)=x(n)-2x(n)
d) Accumulator
Answer: d
Explanation: An identity system gives the output same as input hence it totally depends on the
present state of the input. Therefore, it is memory less. Similarly, a resistor and the expression in
option c are memory less systems as they depend upon the present state of the input. An
accumulator sums up the values of all past and present states of input. Therefore, it is a system
with memory.

Department of Electrical and Computer Engineering 26

Wollo University Kombolcha Institute of Technology

108. Which among the following is a memory less system?

a) Delay
b) Summer
c) Resistor
d) Capacitor
Answer: c
Explanation: Options Delay, Summer and Capacitor are all systems with memory as they depend
upon past, past and present, past and present values of input respectively. Whereas, a resistor is a
memory less system as its relationship with output always depends upon the current or present
state of the input.

109. In a continuous-time physical system, memory is directly associated with _____________

a) Storage registers
b) Time
c) Storage of energy
d) Number of components in the system
Answer: c
Explanation: Memory is directly associated with storage of energy such as electric charge in the
capacitor or kinetic energy in an automobile. Storage registers are for discrete time systems such
as microprocessor etc. Time and number of components of a system have got nothing to do with
memory.

110. A system with memory which anticipates future values of input is called _________
a) Non-causal System
b) Non-anticipative System
c) Causal System
d) Static System
Answer: a
Explanation: A system which anticipates the future values of input is called a non-causal system.
A causal depends only on the past and present values of input. Non-anticipative is another name
for the causal system. A static system is memory less system.

111. Determine the nature of the system: y(n)=x(-n).

a) Causal
b) Non-causal
c) Causal for all positive values of n
d) Non-causal for negative values of n
Answer: b
Explanation: The given system gives negative values of input i.e., past values of input when we
feed positive integers to LHS. However, it gives positive values for negative values of n i.e.,
future values. Therefore, the system depends upon past values for some integers and future
values for some other. A system cannot be called partially causal or non-causal, therefore, the
system is non-causal.

112. Which among the following is an application of non-causal system?

a) Image processing

Department of Electrical and Computer Engineering 27

Wollo University Kombolcha Institute of Technology

b) RC circuit
c) Stock market Analysis
d) Automobile
Answer: c
Explanation: Image processing, RC circuit, and an automobile are all causal systems as they do
not anticipate the future values of an image, RC circuit and future actions of a driver
respectively. Instead, they function upon either the stored information or on the current values of
the input. Whereas, in the stock market, analysts try to figure out a trend in the future based upon
the stored information. Therefore, it is non-causal.

113. Determine the nature of the given system: y(t)=x(sint)

a) Causal, Non-linear
b) Causal, Linear
c) Non-Causal, Non-linear
d) Non-causal, Linear
Answer: d
Explanation: The system is non-causal as it gives future values for some inputs.
E.g. y (- π) = x (sin (-π)) = x (0)
For linearity, it needs to satisfy superposition principle,
⇒ y1 (t) = x1 (sint)
⇒ y2 (t) = x2 (sint)
⇒ ay1 (t) + by2 (t) = ax1 (sint) + bx1 (sint) Equation 1
Now, y3 (t) = x3 (sint) = (ax1 + bx2)(sint) = ax1 (sint) + bx1 (sint) Equation 2
Clearly, Equation 1 and 2 are equal, hence the system is linear.

114. Is the system y[n]=2x[n]+2 linear?

a) YES
b) NO
Answer: b
Explanation: The system needs to satisfy superposition principle for linearity:
For input x1[n], y1 [n] = 2x1 [n] + 2
For input x2[n], y2 [n] = 2x2 [n] + 2
⇒ ay1 [n]+ by2 [n] = 2(ax1 [n]+ bx2 [n]) + 2(a+b) Equation 1
For, x3[n], y3 [n]=2x3 [n]+2 = 2(ax1 [n]+ bx2 [n]) + 2 Equation 2
Clearly, Equation 1 is not equal to equation
∴ The system does not satisfy superposition principle ⇒ The system is not linear.

115. An inverse system with the original system gives an output equal to the input. How is the
inverse system connected to the original system?
a) Series
b) Cascaded
c) parallel
d) No connection
Answer: c
Explanation: An inverse system when cascaded with the original system gives an output equal to
the input.

Department of Electrical and Computer Engineering 28

Wollo University Kombolcha Institute of Technology

116. Which among the following is an invertible system?

a) y[n] = 0
b) y[n] = 2x[n]
c) y(t) = x2(t)
d) y(t) = dx(t)/dt
Answer: b
Explanation: A system is said to be invertible if it’s input can be found out from its output.
Implying, if a system has same outputs for several inputs then it is impossible to find the correct
input as output is same for many. Therefore, a system is invertible if it gives distinct outputs to
distinct inputs. It is non-invertible if it gives same outputs for many inputs.
Option a produces 0 output for any input → Non-invertible
Option b produces different outputs for different inputs and also it’s inverse system is (1/2)y[n]
→ Invertible
Option c, we get same output for both positive and negative values → Non-invertible
Option d, we get 0 for all constant input values → Non-invertible.

117. Is the system time invariant: y(t) = x(4t)?

a) YES
b) NO
Answer: b
Explanation: A system is said to be time invariant if a change input causes the same change in
output.
For change in input by T
⇒ y(t, T) = x(4(t – T)) = x(4t – 4T) Equation 1
For the same change in output
⇒ y(t – T) = x(4t – T) Equation 2
Equation 1 is not equal to equation 2.
∴ The system is not time invariant or is time variant.

118. Determine the nature of the system: y[n] = x[n]x[n – 1] with unit impulse function as an
input.
a) Dynamic, output always zero, non-invertible
b) Static, output always zero, non-invertible
c) Dynamic, output always 1, invertible
d) Dynamic, output always 1, invertible
Answer: a
Explanation: Since the system depends on present and past values, therefore, it is not memory
less(dynamic).
Now, input is a unit impulse function. Unit impulse function = 1 at n = 0, otherwise it is equal to
0.
For, y[0] = x[0]x[-1] = 1 × 0 = 0
For, y[1] = x[1]x[0] = 0 × 1 = 0
For, y[2] = x[2]x[1] = 0 × 0 = 0
∴ For any time, output is always zero.
Since, the output is always same, the system is non-invertible

Department of Electrical and Computer Engineering 29

Wollo University Kombolcha Institute of Technology

119. What is a stable system?

a) If every bounded input results in the bounded output
b) If every bounded input results in an unbounded output
c) If every unbounded input results in a bounded output
d) If unbounded input results in bounded as well as unbounded output
Answer: a
Explanation: The system is said to bounded input bounded output stable if every bounded input
results in bounded output and also the output of such a system does not diverge if the input does
not diverge.

120. If This is an example for _______ system.

a) Stable system
b) Unstable system
c) Bounded input unbounded output system
d) Unbounded input system
Answer: a
Explanation: In the above example, the input is finite and the output is also finite as y (t) is
absolute integrable. Hence the system is stable.

121. If x(t)= ∂(t-1) and y(t)= e-t. This is an example for ______ system.
a) Stable
b) BIBO
c) Bounded input
d) Unstable
Answer: d
Explanation: In this example, the input is finite and output is not finite. Hence the given system
is unstable.

122. If x(t)=et, y(t)= e-2t this is a _____system.

a) Unstable
b) Stable
c) BIBO
d) Cannot classify the system
Answer: d
Explanation: In this example, the input is infinite and hence this input cannot be used to classify
the system. Here the output is not considered.

123. Which of the following is not true about systems having memory?
a) It is also called dynamic systems
b) The output signal depends on the past values of the input signal
c) It is also called static system
d) Resistive circuit
Answer: c
Explanation: The system is said to have memory if its output signal depends on the past values of
the input signal and also it is called dynamic system.
Department of Electrical and Computer Engineering 30
Wollo University Kombolcha Institute of Technology

124. How far does the memory of the given system y[n]=1/2{x[n]+ x[n-1]} extend into past?
a) Two time units
b) One time unit
c) Three time units
d) Not predictable
Answer: b
Explanation: The given memory system extends into past by one time unit. It is determined by
the term x [n-1].

125. The input- output relation of a device is represented asi(t)=ao+a1v1(t)+a2v2 (t)+⋯. Does
this device have memory?
a) Has memory
b) Does not have memory
c) It is dynamic
d) Insufficient information
Answer: b
Explanation: In the given equation, v (t) is the voltage applied I (t) is the current flowing through
the device and a0, a1, a2 are the constants. It does not involve any past value of the input signal
and hence memory less.

126. Which is not an example for memory system?

a) Capacitive circuit
b) Inductive circuit
c) Resistive circuit
d) Parallel RC circuit
Answer: c
Explanation: Resistive circuit is memory less since the current I (t) flowing through it in
response to the applied voltage v (t) is defined by i(t) = 1⁄R v(t).

127. What is the memory of the system if its input-output relation is given by

?
a) Memory extends from time t to the infinite future
b) Memory extends from time t to the infinite past
c) Does not have memory
d) Insufficient information
Answer: b
Explanation: Given system has inductor involved in it. Hence it has memory. Since integral is
from -∞ to time t, its memory extends from time t to infinite past.

128. Which of the following systems is memory less?

a) y(t) = 2x(t) + d⁄dx x(t)
b) y(t) = 2x2 (t) + d⁄dx x(t)
c) y(t) = ∫x(t)dt
d) y(t) = 2x2 (t)
Answer: d

Department of Electrical and Computer Engineering 31

Wollo University Kombolcha Institute of Technology

Explanation: A differentiator or integrator maybe realized with capacitors and inductors and
cannot be realized using resistors. Hence differentiators and integrators can be considered as
systems with memory.

129. An example for non-causal system is ________

a) Amplifier
b) Oscillator
c) Rectifiers
d) Does not exists
Answer: d
Explanation: Non-causal system is the one which results in output even without the application
of input. Since all systems are real, non-causal systems do not exists.

130. Is Ideal low pass filter is an example for Non –causal system?
a) True
b) False
Answer: a
Explanation: Ideal low pass filter has sharp transitions which cannot be physically realized.
Hence non – causal.

131. Can impulse response be measured?

a) Impulse cannot be generated
b) Impulse can be generated
c) Can be measured
d) Cannot be measured
Answer: c
Explanation: Impulse response can be measured but in an indirect manner. Hence by giving step
response to a suitable differentiator impulse response is measured. Impulse response is derivative
of step response.

132. Which of the following is an example for non- causal system?

a) y[n] = 1⁄3 {x[n-1] + x[n] + x[n-2]}
b) y[n] = 1⁄3 {x[n-1] + x[n] + x[n+1]}
c) y[n] = 1⁄2 {x[n-1] + x[n]}
d) y[n] = 1⁄2 {x[n] + x[n-2]}
Answer: b
Explanation: y[n] = 1⁄3 {x[n-1] + x[n] + x[n+1]} is an example for non- causal system since the
output y [n] depends on the future value of the input namely x [n+1].

133. Which of the following is not true about invertible systems?

a) H-1 H=I
b) There must be one-to-one mapping between input and output signals for a system to be
invertible
c) Input of the invertible system can be recovered from the system output
d) Input of the invertible system cannot be recovered from the system output
Answer: d

Department of Electrical and Computer Engineering 32

Wollo University Kombolcha Institute of Technology

Explanation: By the definition of invertible system we can say that input of the invertible system
can be recovered from the system output.

134. Is y(t)= x2 (t) is an example for invertible system?

a) True
b) False
Answer: b
Explanation: In this example we don’t have a unique inverse hence the input of the system is not
recovered from the system output. Hence it is considered as non-invertible system.

135. y(t) = 2x(t) + 3t d⁄dx x(t) Is an example for _____

a) Time invariant system
b) Time varying system
c) LTI system
d) Time invariant and linear system
Answer: b
Explanation: The given system does not satisfy the condition {R{x(t-to)}=y(t-to)} hence the
system is time varying.

136. y(t) = 5x(t) + 6 d⁄dx x(t) Is an example for _____ system.

a) Time varying
b) Time invariant
c) Time varying and linear
d) Time varying and non linear
Answer: b
Explanation: The given system satisfies the condition {R{x(t-to)}=y(t-to)} hence the system is
time invariant

Discrete Time Signals

137. Is the function y[n] = sin(x[n]) periodic or not?
a) True
b) False
Answer: b
Explanation: ‘y’ will be periodic only if x attains the same value after some time, T. However, if
x is a one-one discrete function, it may not be possible for some x[n].

138. What is the time period of the function x[n] = exp(jwn)?

a) pi/2w
b) pi/w
c) 2pi/w
d) 4pi/w
Answer: c
Explanation: Using Euler’s rule, exp(2pi*n) = 1 for all integer n. Thus, the answer can be
derived.

Department of Electrical and Computer Engineering 33

Wollo University Kombolcha Institute of Technology

139. What is the nature of the following function: y[n] = y[n-1] + x[n]?
a) Integrator
b) Differentiator
c) Subtractor
d) Accumulator
Answer: d
Explanation: If the above recursive definition is repeated for all n, starting from 1,2.. then y[n]
will be the sum of all x[n] ranging from 1 to n, making it an accumulator system.

140. Is the above function defined, causal in nature?

a) True
b) False
Answer: a
Explanation: As the value of the function depends solely on the value of the input at a time
presently and/or in the past, it is a causal system.

141. Is the function y[n] = x[n-1] – x[n-4] memoryless?

a) True
b) False
Answer: b
Explanation: Since the function needs to store what it was at a time 4 units and 1 unit before the
present time, it needs memory.

142. Is the function y[n] = x[n-1] – x[n-56] causal?

a) The system is non causal
b) The system is causal
c) Both causal and non causal
d) None of the mentioned
Answer: b
Explanation: As the value of the function depends solely on the value of the input at a time
presently and/or in the past, it is a causal system.

143. Is the function y[n] = y[n-1] + x[n] stable in nature?

a) It is stable
b) It is unstable
c) Both stable and unstable
d) None of the mentioned
Answer: a
Explanation: It is BIBO stable in nature, i.e. bounded input-bounded output stable.

144. If n tends to infinity, is the accumulator function a stable one?

a) The function is marginally stable
b) The function is stable
c) The function is unstable
d) None of the mentioned
Answer: c

Department of Electrical and Computer Engineering 34

Wollo University Kombolcha Institute of Technology

Explanation: The system would be unstable, as the output will grow out of bound at the
maximally worst possible case.

145. We define y[n] = nx[n] – (n-1)x[n]. Now, z[n] = z[n-1] + y[n], is z[n] stable?
a) Yes
b) No
Answer: a
Explanation: As we take the sum of y[n], terms cancel out and deem z[n] to be BIBO stable.

146. We define y[n] = nx[n] – (n-1)x[n]. Now, z[n] = z[n-1] + y[n]. Is z[n] a causal system?
a) No
b) Yes
Answer: b
Explanation: As the value of the function depends solely on the value of the input at a time
presently and/or in the past, it is a causal system.

147. Discrete-time signals are _________________

a) Continuous in amplitude and continuous in time
b) Continuous in amplitude and discrete in time
c) Discrete in amplitude and discrete in time
d) Discrete in amplitude and continuous in time
Answer: b
Explanation: A discrete-time signal is continuous in amplitude and discrete in time. It can either
be present in nature or is sampled from an analog signal. A digital signal is discrete in amplitude
and time.

148. Determine the discrete-time signal: x(n)=1 for n≥0 and x(n)=0 for n<0
a) Unit ramp sequence
b) Unit impulse sequence
c) Exponential sequence
d) Unit step sequence
Answer: d
Explanation: Unit step is defined by: x(n)=1 for n≥0 and x(n)=0 for n<0.

149. Determine the value of the summation: ∑∞n= -∞δ(n-1)sin2n.

a) 1
b) 0
c) sin2
d) sin4
Answer: c
Explanation: ∑∞n= -∞δ(n-1)sin2n
⇒ We know, δ(n) is impulse function which means δ(n)=1 when n=0
⇒ δ(n-1)=1 when n=1 otherwise it is 0.
Therefore, the summation’s limit reduces to n=1
⇒ ∑∞n= -∞δ(n-1)sin2n = sin2n|n=1 = sin2.

Department of Electrical and Computer Engineering 35

Wollo University Kombolcha Institute of Technology

150. Determine the value of the summation: ∑∞n= -∞ δ(n+3)(n2+n).

a) 3
b) 6
c) 9
d) 12
Answer: b
Explanation: ∑∞n= -∞ δ(n+3)(n2+n)
⇒ δ(n+3)=1 when n= -3 otherwise 0.
Therefore, the limit reduces to n = -3
⇒ ∑∞n = -∞ δ(n+3)(n2+n) = (n2+n)|n = -3 = (-3)2-3 = 9 – 3 = 6.

151. Determine the product of two signals: x1 (n) = {2,1,1.5,3}; x2 (n) = { 1,1.5,0,2}.
a) {2,1.5,0,6}
b) {2,1.5,6,0}
c) {2,0,1.5,6}
d) {2,1.5,0,3}
Answer: a
Explanation: Product of discrete-time signals is computed element by element.
⇒ x(n) = x1 (n) * x2 (n) = {2×1, 1×1.5, 1.5×0, 3×2} = {2,1.5,0,6}

Useful Signals
152. What is the value of d[0], such that d[n] is the unit impulse function?
a) 0
b) 0.5
c) 1.5
d) 1
Answer: d
Explanation: The unit impulse function has value 1 at n = 0 and zero everywhere else.

153. What is the value of u[1], where u[n] is the unit step function?
a) 1
b) 0.5
c) 0
d) -1
Answer: a
Explanation: The unit step function u[n] = 1 for all n>=0, hence u[1] = 1.

154. Evaluate the following function in terms of t: {sum from -1 to infinity:d[n]}/{Integral from
0 to t: u(t)}
a) t
b) 1⁄t
c) t2
d) 1⁄t2

Department of Electrical and Computer Engineering 36

Wollo University Kombolcha Institute of Technology

Answer: b
Explanation: The numerator evaluates to 1, and the denominator is t, hence the answer is 1/t.

155. Evaluate the following function in terms of t: {integral from 0 to t}{Integral from -inf to
inf}d(t)
a) 1⁄t
b) 1⁄t2
c) t
d) t2
Answer: c
Explanation: The first integral is 1, and the overall integral evaluates to t.

156. The fundamental period of exp(jwt) is

a) pi/w
b) 2pi/w
c) 3pi/w
d) 4pi/w
Answer: b
Explanation: The function assumes the same value after t+2pi/w, hence the period would be
2pi/w.

157. Find the magnitude of exp(jwt). Find the boundness of sin(t) and cos(t).
a) 1, [-1,2], [-1,2]
b) 0.5, [-1,1], [-1,1]
c) 1, [-1,1], [-1,2]
d) 1, [-1,1], [-1,1]
Answer: d
Explanation: The sin(t)and cos(t) can be found using Euler’s rule.

158. Find the value of {sum from -inf to inf} exp(jwn)*d[n].

a) 0
b) 1
c) 2
d) 3
Answer: b
Explanation: The sum will exist only for n = 0, for which the product will be 1.

159. Compute d[n]d[n-1] + d[n-1]d[n-2] for n = 0, 1, 2.

a) 0, 1, 2
b) 0, 0, 1
c) 1, 0, 0
d) 0, 0, 0
Answer: d
Explanation: Only one of the values can be one at a time, others will be forced to zero, due to the
delta function.

Department of Electrical and Computer Engineering 37

Wollo University Kombolcha Institute of Technology

160. Defining u(t), r(t) and s(t) in their standard ways, are their derivatives defined at t = 0?
a) Yes, Yes, No
b) No, Yes, No
c) No, No, Yes
d) No, No, No
Answer: d
Explanation: None of the derivatives are defined at t=0.

161. Which is the correct Euler expression?

a) exp(2jt) = cos(2t) + jsin(t)
b) exp(2jt) = cos(2t) + jsin(2t)
c) exp(2jt) = cos(2t) + sin(t)
d) exp(2jt) = jcos(2t) + jsin(t)
Answer: b
Explanation: Euler rule: exp(jt) = cos(t) + jsin(t).

162. The range for unit step function for u(t – a), is ________
a) t < a
b) t ≤ a
c) t = a
d) t ≥ a
Answer: d
Explanation: A unit step signal u(t) = 1 when t ≥ 0 and 0 when t < 0
∴ u(t – a) = 1 when t – a ≥ 0 ⇒ t ≥ a

163. Which one of the following is not a ramp function?

a) r(t) = t when t ≥ 0
b) r(t) = 0 when t < 0
c) r(t) = ∫u(t)dt when t < 0
d) r(t) = du(t)⁄dt
Answer: d
Explanation: Ramp function r(t) = t when t ≥ 0 and r(t) = 0 when t < 0
Also, r(t)= ∫u(t)dt = ∫dt = t (∵u(t) = 1 for t≥0)
⇒ du(t)⁄dt = d(1)⁄dt = 0 which is not a ramp function.

164. Which one of the following is not a unit step function?

Answer: d

Department of Electrical and Computer Engineering 38

Wollo University Kombolcha Institute of Technology

Explanation: Unit step function, u(t) = 1 for t ≥ 0 and u(t) = 0 for t < 0. Also,

165 Unit Impulse function is obtained by using the limiting process on which among the
following functions?
a) Triangular Function
b) Rectangular Function
c) Signum Function
d) Sinc Function
Answer: b
Explanation: Unit impulse function can be obtained by using a limiting process on the
rectangular pulse function. Area under the rectangular pulse is equal to unity.

166. Evaluate:
a) {2,1.5,0,6}
b) {2,1.5,6,0}
c) {2,0,1.5,6}
d) {2,1.5,0,3}
Answer: a
Explanation: From the impulse function property,

167. When is a complex exponential signal pure DC?

a) σ = 0 and Ω < 0
b) σ < 0 and Ω = 0
c) σ = 0 and Ω = 0
d) σ < 0 and Ω < 0
Answer: c
Explanation: A complex exponential signal is represented as x(t)= est
Where, s = σ + jΩ
⇒ x(t) = eσt [cosΩt + jsinΩt] When, σ = 0 and Ω = 0 ⇒ x(t) = e0 [cos0 + jsin0] = 1 × 1 = 1 which
is pure DC.

Department of Electrical and Computer Engineering 39

Wollo University Kombolcha Institute of Technology

Discrete-Time Systems in the Time-Domain

168. Is the function y[n] = cos(x[n]) periodic or not?
a) True
b) False
Answer: a
Explanation: ‘y’ will be periodic only if x attains the same value after some time, T. However, if
x is a one-one discrete function, it may not be possible for some x[n].

169. If n tends to infinity, is the accumulator function an unstable one?

a) The function is marginally stable
b) The function is unstable
c) The function is stable
d) None of the mentioned
Answer: b
Explanation: The system would be unstable, as the output will grow out of bound at the
maximally worst possible case.

170. Comment on the causality of the following discrete time system: y[n] = x[-n].
a) Causal
b) Non causal
c) Both Casual and Non casual
d) None of the mentioned
Answer: b
Explanation: For positive time, the output depends on the input at an earlier time, giving
causality for this portion. However, at a negative time, the output depends on the input at a
positive time, i.e. at a time in the future, rendering it non causal.

171. Comment on the causality of the discrete time system: y[n] = x[n+3].
a) Causal
b) Non Causal
c) Anti Causal
d) None of the mentioned
Answer: c
Explanation: The output always depends on the input at a time in the future, rendering it anti-
causal.

172. Consider the system y[n] = 2x[n] + 5. Is the function linear?

a) Yes
b) No
Answer: b
Explanation: As we give two inputs, x1 and x2, and give an added input x1 and x2, we do not get
the corresponding y1 and y2. Thus, additive rule is disturbed and hence the system is not linear.

Department of Electrical and Computer Engineering 40

Wollo University Kombolcha Institute of Technology

173.Comment on the time invariance of the following discrete system: y[n] = x[2n+4].
a) Time invariant
b) Time variant
c) Both Time variant and Time invariant
d) None of the mentioned
Answer: b
Explanation: A time shift in the input scale gives double the time shift in the output scale, and
hence is time variant.

174. Is the function y[2n] = x[2n] linear in nature?

a) Yes
b) No
Answer: a
Explanation: The function obeys both additivity and homogeneity properties. Hence, the function
is linear.

175. How is a linear function described as?

a) Zero in Finite out
b) Zero in infinite out
c) Zero in zero out
d) Zero in Negative out
Answer: c
Explanation: The system needs to give a zero output for a zero input so as to conserve the law of
additivity, to ensure linearity.

176. Is the system y[n] = x2[n-2] linear?

a) Yes
b) No
Answer: b
Explanation: The system is not linear, as x12 + x22 is not equal to (x1 + x2)2.

177. Is the above system, i.e y[n] = x2[n-2] time invariant?

a) Yes
b) No
Answer: a
Explanation: A time shift of t0 will still result in an equivalent time shift of t0 in the output, and
hence will be time invariant

178. The difference equation for an Nth order discrete-time system is ___________

Department of Electrical and Computer Engineering 41

Wollo University Kombolcha Institute of Technology

Answer: c
Explanation: The difference equation for an Nth order discrete-time system is:

179. The response of any discrete time system can be decomposed as _____________
a) Total Response=Impulse+step
b) Total Response=Impulse+Ramp
c) Total Response=zero-output response
d) Total Response=zero-state response+zero-input response
Answer: d
Explanation: There are two approaches to analyzing response of a system:
Direct solution of difference solution
Decomposing in terms of impulse signals
In the first method, the response of the system can be decomposed as:
Total Response = zero-state response + zero-input response.

180. Zero-state response of the system is _____________

a) Response of the system when initial state of the system is zero
b) Response of the system due to input alone
c) Response of the system due to input alone when initial state of the system is zero
d) Response of the system due to input alone when initial state is neglected
Answer: c
Explanation: Zero-state response of the system is the response of the system due to input alone
when the initial state of the system is zero. That is the system is relaxed at time n = 0.

181. Zero-input response is also known as ____________

a) zero-state response
b) Natural response
c) state-input response
d) Forced response
Answer: b
Explanation: Natural response of the system is when the input x(n) = 0.

182. The general solution of natural response is of the form of _________

a) yh (n)= c1 λ1n+c2λ2n+⋯+cNλNn
b) yh (n)= c1 λ1n+c2λ2n+⋯+cNλNn
c) yh (n)= c1 λ12+c2λ22+⋯+cNλN2
d) yh(n)= c1 λ1n-c2λ2n+⋯+cNλNn
Answer: a
Explanation: The general solution of natural response is of the form:
yh (n)= c1 λ1n+c2λ2n+⋯+cNλNn
The form will vary if the roots are repeating or complex.

Department of Electrical and Computer Engineering 42

Wollo University Kombolcha Institute of Technology

183. Determine the natural response of the system: Difference equation is

y(n)-y(n-1)-2y(n-2)=x(n) and y(-1) = 1; y(-2) = 0

Answer: c
Explanation: Natural Response of the system:
Homogenous equation ⇒ y(n)-y(n-1)-2y(n-2)=0
The homogenous solution: yh(n)= λn
⇒ λn– λ(n-1)-2λ(n-2)=0
⇒ λ(n-2) [λ2– λ1-2]=0
⇒ λ2– λ-2=0
⇒ λ2-2λ+λ-2=0
⇒ λ(λ-2)+1(λ-2)=0
⇒ (λ-2)(λ+1)=0
⇒ λ1=2,λ2=-1
General form of homogenous solution is
yh (n)= c1 (2)n+c2(-1)n (1)
⇒ y(0)= c1+c2 (2)
⇒ y(1)=2c1– c2 (3)

⇒ y(0)-y(-1)-2y(-2)=0
Given, y(-1) = 1 and y(-2) = 0
⇒ y(0)-1=0⇒y(0)=1
Similarly, y(1)-y(0)-2y(-1)=0⇒y(1)=1+2=3
∴ y(0) = 1 and y(1) = 3
Comparing the above values with equations (2) and (3)
⇒ c1+c2=1 and 2c1– c2=3
Solving the two equations we get, c1 = 4/3 and c2 = -1/3

184. Forced Response is solution of difference equation when ____________

a) Input is zero
b) Input is given and initial conditions are zero
c) Natural Response
d) Input is given and initial conditions are non-zero
Answer: b
Explanation: Forced response is solution of difference equation when input is given and initial
conditions are zero. Also known as zero-state response.

Department of Electrical and Computer Engineering 43

Wollo University Kombolcha Institute of Technology

185. Forced response consists of _________

a) Homogenous solution and general solution
b) General solution alone
c) Homogenous solution and particular solution
d) Particular solution alone
Answer: c
Explanation: Forced response consists of homogenous solution and particular solution.

Periodic and Non-Periodic Signals

186. Given the signal
X (t) = cos t, if t<0
Sin t, if t≥0
The correct statement among the following is?
a) Periodic with fundamental period 2π
b) Periodic but with no fundamental period
c) Non-periodic and discontinuous
d) Non-periodic but continuous
Answer: c
Explanation: From the graphs of cos and sin, we can infer that at t=0, the function becomes
discontinuous.
Since, cos 0 = 1, but sin 0 = 0
As 1 ≠ 0, so, the function X (t) is discontinuous and therefore Non-periodic.

187. The fundamental period of the signal X (t) = 10 cos2(10 πt) is __________
a) 0.2
b) 0.1
c) 0.5
d) No fundamental period exists
Answer: b
Explanation: X (t) = 10 cos2 (10 πt)
Since, cos 2t = 2cos2 t – 1
Or, cos2 t = 1+cos2t2
∴ X (t) = 5 + 5 cos 20πt
Now, Y (t) = cos 20πt
Fundamental period of the signal is = 2π20π=110 = 0.1.

188. The even component of the signal X (t) = ejt is _________________

a) Sin t
b) Cos t
c) Sinh t
d) Cosh t
Answer: b
Explanation: Let Xe (t) represents the even component of X (t)
Now, Xe (t) = 12[X (t) + X (-t)]

Department of Electrical and Computer Engineering 44

Wollo University Kombolcha Institute of Technology

= 12[ejt + e-jt]
= cos t.

189. The odd component of the signal X (t) = ejt is _______________

a) Sin t
b) Cos t
c) Sinh t
d) Cosh t
Answer: a
Explanation: Let Xo (t) represents the odd component of X (t)
Now, Xo (t) = 12[X (t) – X (-t)]
= 12[ejt + e-jt]
= sin t.

190. The period of the signal X (t) = 10 sin 5t – 4 cos 9t is _______________

a) 24π35
b) 4π35
c) 2π
d) Non-periodic
Answer: c
Explanation: Period of cos t = 2π
Period of cos at = 2πa
Here, a = 9
So, period of cos 9t = 2π9
Again, Period of sin t = 2π
Period of sin at = 2πa
Here, a = 5
So, period of sin 5t = 2π5
∴ Period of X (t) = LCM [Period of X1 (t), Period of X2 (t)]
∴ Period of X (t) = LCM (2π5,2π9) = 2π.

191. The period of the signal X (t) = 5t – 2 cos 6000 πt is ________________

a) 0.96 ms
b) 1.4 ms
c) 0.4 ms
d) Non-periodic
Answer: d
Explanation: Period of cos t = 2π
Period of cos at = 2πa
Here, a = 6000π
So, period of cos 6000πt = 2π6000π
= 13000
Again, Period of t = indefinite
∴ Period of X (t) = LCM [Period of X1 (t), Period of X2 (t)]
∴ Period of X (t) = LCM (13000, ∞) = Indefinite.

Department of Electrical and Computer Engineering 45

Wollo University Kombolcha Institute of Technology

192. The period of the signal X (t) = 4 sin 6t + 3 sin 3–√t is ________________
a) 2π3 s
b) 2π3√ s
c) 2π s
d) Non-periodic
Answer: d
Explanation: Period of sin t = 2π
Period of sin at = 2πa
Here, a = 6
So, period of sin 6t = 2π6
Again, a = 3–√
So, period of sin 3–√t = 2π3√
∴ Period of X (t) = LCM [Period of X1 (t), Period of X2 (t)]
∴ Period of X (t) = LCM (π3,2π3√) = Indefinite.

193. The period of the signal Z (t) = sin3t + cos 4t is _______________

a) periodic without a definite period
b) periodic with a definite period
c) non- periodic over an interval
d) non-periodic throughout
Answer: b
Explanation: Period of cos t = 2π
Period of cos at = 2πa
Here, a = 4
So, period of cos 4t = 2π4
= π2
Again, Period of sin t = 2π
Period of sin at = 2πa
Here, a = 3
So, period of sin 3t = 2π3
∴ Period of X (t) = LCM [Period of X1 (t), Period of X2 (t)]
∴ Period of X (t) = LCM (2π5,2π4) = definite
Hence Z (t) is periodic with a definite period.

194. The signal X (t) = e-4t u (t) is _______________

a) Power signal with P∞ = 14
b) Power signal with P∞ = 0
c) Energy signal with E∞ = 14
d) Energy signal with E∞ = 0
Answer: c
Explanation: If a signal has E∞ as ∞ and P∞ as a finite value, then the signal is a power signal. If
a signal has E∞ as a finite value and P∞ as ∞, then the signal is an energy signal.
|x (t)| < ∞, E∞ = ∫∞−∞|x(t)|2dt
= ∫∞∞e−4tu(t)dt

Department of Electrical and Computer Engineering 46

Wollo University Kombolcha Institute of Technology

= ∈∞∞e−4tdt=14
So, this is not a power signal but an energy signal.
P∞=limT→∞12T∫T−T|x(t)|2dt=∞.

195. The signal X (t) = ej(2t+π6) is ________________

a) Power signal with P∞ = 1
b) Power signal with P∞ = 2
c) Energy signal with E∞ = 2
d) Energy signal with E∞ = 1
Answer: a
Explanation: If a signal has E∞ as ∞ and P∞ as a finite value, then the signal is a power signal. If
a signal has E∞ as a finite value and P∞ as ∞, then the signal is an energy signal.
|x (t)| = 1, E∞ = ∫∞−∞|x(t)|2dt=∞
So, this is a power signal not an energy signal.
P∞=limT→∞12T∫T−T|x(t)|2dt=1..
196. Signal X (t) is as shown in the figure below.

The total energy of X (t) is _______________

a) 0
b) 13
c) 133
d) 263
Answer: d
Explanation: E = 2∫50x2(t)dt
= 2 ∫4011dt+2∫54(5–t2)dt
= 8 + 23=263.

197. A discrete time signal is as given below

X[n]=cosπn9+sin(πn7+12)
The period of the signal X [n] is ______________
a) 126

Department of Electrical and Computer Engineering 47

Wollo University Kombolcha Institute of Technology

b) 32
c) 252
d) Non-periodic
Answer: a
Explanation: Given that, N1 = 18, N2 = 14
We know that period of X [n] (say N) = LCM (N1, N2)
∴ Period of X [n] = LCM (18, 14) = 126.

198. A discrete time signal is as given below

X[n]=cos(n8)cos(πn8)
The period of the signal X [n] is _____________
a) 16 π
b) 16(π+1)
c) 8
d) Non-periodic
Answer: d
Explanation: We know that for X [n] = X1 [n] × X2 [n] to be periodic, both X1 [n] and X2 [n]
should be periodic with finite periods.
Here X2 [n] = cos (πn8), is periodic with fundamental period as 8n
But X1 [n] = cos (n8) is non periodic.
∴ X [n] is a non-periodic signal.

199. A discrete time signal is as given below

X[n]=cos(πn2)–sin(πn8)+3cos(πn4+π3)
The period of the signal X [n] is _____________
a) 16
b) 4
c) 2
d) Non-periodic
Answer: a
Explanation: Given that, N1 = 4, N2 = 16, N3 = 8
We know that period of X [n] (say N) = LCM (N1, N2, N3)
∴ Period of X [n] = LCM (4, 16, 8) = 16

Periodic Signals
200. What are periodic signals?
a) The signals which change with time
b) The signals which change with frequency
c) The signal that repeats itself in time
d) The signals that repeat itself over a fixed frequency
Answer: c
Explanation: Those signals which repeat themselves in a fixed interval of time are called periodic
signals. The continuous-time signal x(t) is periodic if and only if
x(t+T)= x(t).
Department of Electrical and Computer Engineering 48
Wollo University Kombolcha Institute of Technology

201. Periodic signals are different in case of continuous time and discrete time signals.
a) True
b) False
Answer: b
Explanation: Periodic signals are same in case of continuous time and discrete time signals.
In case of continuous time signal, x(t)=x(t+T), for all t>0
In case of discrete time signal,
x(n)=x(n+N), for all n>0.

202. What is the time period of a periodic signal in actual terms?

a) The signals which start at t=-∞ and end at t=+∞
b) The signals which have a finite interval of occurrence
c) The signals which start at t= -∞ and ends at a finite time period
d) The signals which have a short period of occurrence
Answer: a
Explanation: The periodic signals have actually a time period between t=-∞ and at t= + ∞. These
signals have an infinite time period, that is periodic signals are actually continued forever. But
this is not possible in case of real time signals.

203. Periodic signals actually exist according to a definition.

a) True
b) False
Answer: b
Explanation: Periodic signals are defined as signals having time period in between t=-∞ and t=+
∞. These signals have an infinite time period that is periodic signals are continued forever. But
real time signals always cease at some time due to distortion and resistance.

204. What is a fundamental period?

a) Every interval of a periodic signal
b) Every interval of an aperiodic signal
c) The first interval of a periodic signal
d) The last interval of a periodic signal
Answer: c
Explanation: The first time interval of a periodic signal after which it repeats itself is called a
fundamental period. It should be noted that the fundamental period is the first positive value of
frequency for which the signal repeats itself.

205. Comment on the periodicity of a constant signal?

a) It is periodic
b) It is not periodic
c) It is a mixture of period and aperiodic signal
d) It depends on the signal
Answer: b
Explanation: A constant signal is not periodic. It is because it does not repeat itself over in time.
It is constant at any time, it is aperiodic.

Department of Electrical and Computer Engineering 49

Wollo University Kombolcha Institute of Technology

206. A discrete time periodic signal is defined as x(n)= x(n+N)

How is the N defined here?
a) Samples/ cycle
b) Samples/ twice cycle
c) Fundamental period
d) Rate of change of the period
Answer: a
Explanation: The value of N is a positive integer and it represents the period of any discrete time
periodic signal measured in terms of number of sample spacing ( samples/cycle). The smallest
value of N is a fundamental period.

207. What is the general range of a period of a signal?

a) It can have of any value from positive to negative
b) It can be negative
c) It can be positive
d) It is always positive
Answer: d
Explanation: The period of a periodic signal is always positive. The smallest positive value of a
periodic interval is called a fundamental period in case of both discrete and continuous time
signal.

208. What is the area of a periodic signal in a periodic interval?

a) It depends on the situation
b) It is same as the area in the previous interval
c) It is different in different situations
d) It is the square of the fundamental period
Answer: b
Explanation: The area of any periodic signal in any interval is the same. Hence it is same as the
previous interval. This results from the fact that a periodic signal takes same values at the
intervals of T.

209. When is the sum of M periodic signals periodic?

a) T/Ti = 1
b) T/Ti = 4
c) T/Ti = ni
d) T/Ti = m+n
Answer: c
Explanation: The sum of M periodic signal is not necessarily periodic. It is periodic only with the
condition that
T/Ti = ni, 1≤i≤M,
where Ti is the period of the signal and in the sum of ni is an integer.

210. How is the period of the sum signal computed as?

a) T*n
b) T*T
c) T*N+M

Department of Electrical and Computer Engineering 50

Wollo University Kombolcha Institute of Technology

d) T *(n+m)
Answer: a
Explanation: If a signal is periodic then we have to convert each of the periods to the ratio of
integers. We have to take the ratio of greatest common divisor(gcd) from the numerator to the
gcd of denominator. The LCM of the denominators of the resulting ratios is the value of n the
period of the sum signal is T*n.

211. What is the necessary and sufficient condition for a sum of a periodic continuous time
signal to be periodic?
a) Ratio of period of the first signal to period of other signals should be constant
b) Ratio of period of the first signal to period of other signals should be finite
c) Ratio of period of the first signal to period of other signals should be real
d) Ratio of period of first signal to period of other signal should be rational
Answer: d
Explanation: The necessary and sufficient condition for a sum of a periodic continuous time
signal to be periodic is that the ratio of a period of the first signal to the period of other signals
should be rational.
I.e T/Ti = a rational number

212. Under what conditions the three signals x(t), y(t) and z(t) with period t1 t2 and t3
respectively are periodic?
a) t1/t2= t2/t3
b) t1/t2 is rational
c) All the ratios of the three periods in any order is rational
d) t1/t2/t3= rational
Answer: c
Explanation: if x(t) , y(t) and z(t) are to be periodic then,
t1/t2 should be rational and simultaneously
t1/t3 should be rational and
t2/t3 should be rational. Hence, all the ratios of the three periods in any order is rational.

213. What is the fundamental period of the signal : ejwt?

a) 2π/w
b) 2π/w2
c) 2π/w3
d) 4π/w
Answer: a
Explanation: The complex exponential signal can be represented as
ejwt= ejwt+jwT
Hence, wt=2 π,
T= 2π/w.

214. What is the period of the signal :jejw11t?

a) 2π/10
b) 2π/11
c) 4π/10

Department of Electrical and Computer Engineering 51

Wollo University Kombolcha Institute of Technology

d) 4π/11
Answer: b
Explanation: From the definition of periodic signal, we express a periodic exponential signal as :
ejw11t= ejwt+jwT
Hence, 11wt=2 π,
which gives the fundamental period as
2π/11.

215. Is the sum of discrete time periodic signals periodic?

a) No, they are not
b) Yes they are
c) Depends on the signal
d) Not periodic if their ratio is not rational
Answer: b
Explanation: The sum of discrete time periodic signals always periodic because the period ratios
N/N are always rational.
For the continuous time, it depends on the ratio.

216. How can we generate a periodic signal from a periodic signal itself?
a) By extending a signal with duration T
b) Cannot be extended
c) By extending the periodic signal’s amplitude
d) By extending the sugar with duration 2π
Answer: a
Explanation: A periodic signal x(t) can be generated by a periodic extension of any segment of x
of duration T( the period).
As a result, we can generate x(t) from any segment x(t) having a duration of one period by
replacing this segment and reproduction thereof end to end ad infinitum on either side.

217. Is a non periodic signal same as aperiodic signal?

a) No, it is not same as an aperiodic signal
b) Yes it is the other name of aperiodic signal
c) It is a branch of aperiodic signal
d) Aperiodic signal is a branch of non periodic signal
Answer: b
Explanation: A signal which does not satisfy the condition:
x(t) = x(t+T) is called an aperiodic signal.
Non periodic is another name of an aperiodic signal. Hence it is exactly the same.

218. What is the period of the signal: 2cost/6?

a) 8π
b) 16π
c) 12π
d) 10π
Answer: c
Explanation: Comparing the above signal with the standard form Acos2πFt, where A is the

Department of Electrical and Computer Engineering 52

Wollo University Kombolcha Institute of Technology

amplitude and F is the frequency,

We get, 2πF=⅙
So, F= 1/12π
Hence, t= 12π.

219. When a continuous signal is a mixture of two continuous periodic signals, what is its
periodicity?
a) LCM of the periods of the two signals, provide their ratio is unity
b) LCM of the periods of the two signals, provide their ratio is rational
c) HCF of the periods of the two signals, provide their ratio is rational
d) LCM of the periods of the two signals provide their ratio is real
Answer: b
Explanation: When a continuous signal is a mixture of two continuous periodic signals if their
time periods are T1 and T2, and their ratio is rational number, then, the periodicity of the
continuous time signal will be the LCM of T1 and T2.

220. Is the signal eαt periodic?

a) Not periodic
b) Yes periodic
c) Depends on the value of
d) Semi- periodic
Answer: c
Explanation: Using the definition of x(t),
x(t) = eαt
ejwt = ejwt+jwαT
For any value of α, if alpha is positive, it has a remaining term ejwαT
Hence it is not periodic.

221. What is a fundamental angular frequency?

a) The inverse of the fundamental time period
b) The inverse of fundamental frequency
c) Fundamental frequency in radians
d) Fundamental frequency in degree
Answer: c
Explanation: The inverse of the fundamental time period is called fundamental frequency. If it is
F, then 2πF is called the fundamental angular frequency ie it is a fundamental frequency in
radians.

222. What is the period of cos3t + sin14t?

a) 4π
b) π
c) 2π
d) 3π
Answer: b
Explanation: We know, T1= 2 π/3 and T2= 2 π/14
Now, T1/T2=14/3.

Department of Electrical and Computer Engineering 53

Wollo University Kombolcha Institute of Technology

So, LCM gives the time period as π.

223. What is the periodicity of a discrete time signal?

a) 2πm/w
b) 2πm/w
c) 2πm/w
d) 2πm/w
Answer: b
Explanation: Using exponential function, we can show that
2π/N= w/m
Which when rearranged gets us 2πm/w.

224. What is the condition of a periodicity of exponential signal eαt?

a) α=1
b) α=2
c) α=3
d) Depends on equation
Answer: a
Explanation: From, x(t+T)= eαt+T = eαt eαt. For any value of α, eαt ≠1 so x(t+T) ≠x(t). So only if
α=1, the signal will be periodic

The Impulse Function

225. How is the discrete time impulse function defined in terms of the step function?
a) d[n] = u[n+1] – u[n].
b) d[n] = u[n] – u[n-2].
c) d[n] = u[n] – u[n-1].
d) d[n] = u[n+1] – u[n-1].
Answer: c
Explanation: Using the definition of the Heaviside function, we can come to this conclusion.

226. What is the definition of the delta function in time space intuitively?
a) Defines that there is a point 1 at t=0, and zero everywhere else
b) Defines that there is a point 0 at t=0, and 1 everywhere else
c) Defines 1 for all t > 0, and 0 else
d) Defines an impulse of area 1 at t=0, zero everywhere else
Answer: d
Explanation: Arises from the definition of the delta function. There is a clear difference between
just the functional value and the impulse area of the delta function.

227. Is it practically possible for us to provide a perfect impulse to a system?

a) Certainly possible
b) Impossible
c) Possible
d) None of the mentioned

Department of Electrical and Computer Engineering 54

Wollo University Kombolcha Institute of Technology

Answer: b
Explanation: The spread of the impulse can never be restricted to a single point in time, and thus,
we cannot achieve a perfect impulse.

228. The convolution of a discrete time system with a delta function gives
a) the square of the system
b) the system itself
c) the derivative of the system
d) the integral of the system
Answer: b
Explanation: The integral reduces to the the integral calculated at a single point, determined by
the centre of the delta function.

229. Find the value of 2sgn(0)d[0] + d[1] + d[45], where sgn(x) is the signum function.
a) 2
b) -2
c) 1
d) 0
Answer: d
Explanation: sgn(0)=0, and d[n] = 0 for all n not equal to zero. Hence the sum reduces to zero.

230. Where h*x denotes h convolved with x, x[n]*d[n-90] reduces to

a) x[n-89].
b) x[n-91].
c) x[n=90].
d) x[n].
Answer: c
Explanation: The function gets shifted by the center of the delta function during convolution.

231. Where h*x denotes h convolved with x, find the value of d[n]*d[n-1].
a) d[n].
b) d[n-1].
c) d2[n].
d) d2[n-1].
Answer: b
Explanation: Using the corollary, if we take d[n] to be the ‘x’ function, it will be shifted by -1
when convolved with d[n-1], thus rendering d[n-1].

232. How is the continuous time impulse function defined in terms of the step function?
a) u(t) = d(d(t))/dt
b) u(t) = d(t)
c) d(t) = du/dt
d) d(t) = u2(t)
View AnswerAnswer: c
Explanation: Using the definition of the Heaviside function, we can come to this conclusion.

Department of Electrical and Computer Engineering 55

Wollo University Kombolcha Institute of Technology

233. In which of the following useful signals, is the bilateral Laplace Transform different from
the unilateral Laplace Transform?
a) d(t)
b) s(t)
c) u(t)
d) all of the mentioned
Answer: c
Explanation: The bilateral LT is different from the aspect that the integral is applied for the entire
time axis, but the unilateral LT is applied only for the positive time axis. Hence, the u(t) [unit
step function] differs in that aspect and hence can be used to differentiate the same.

234. What is the relation between the unit impulse function and the unit ramp function?
a) r = dd(t)/dt
b) d = dr/dt
c) d = d2(r)/dt2
d) r = d2(d)/dt2
Answer: c
Explanation: Now, d = du/dt and u = dr/dt. Hence, we obtain the above answer.

235. What is the other name of a Continuous Time Unit Impulse Function?
a) Dirac delta function
b) Unit function
c) Area function
d ) Direct delta function
Answer: a
Explanation: The continuous time unit impulse function is also known as the Dirac delta
function. This because it was first defined by Paul Adrein Maurice Dirac as ∂(t)=0.

236. What is the area of a Unit Impulse function?

a) Zero
b) Half of Unity
c) Depends on the function
d) Unity
Answer: d
Explanation: The area under an impulse function is unity. It is defined between limits negative
infinity to positive infinity with ∂(t)dt=1, i.e ∫∂(t)dt=1. It can be seen as a rectangular pulse with
width that is negligible and the height that is infinitely large and area as one.

237. Why is the impulse duration important?

a) It is zero
b) It changes with time
c) It approaches zero
d) It depends on the situation
Answer: c
Explanation: One of the most interesting features of the impulse function, is not its shape, but the

Department of Electrical and Computer Engineering 56

Wollo University Kombolcha Institute of Technology

fact that its effective duration (pulse width) approaches zero, while the area remains unity.
Hence, ∫∂(t)dt=1.

238. What are the singularity functions?

a) Derivatives and integrals of unit impulse functions
b) Derivatives of a unit impulse function
c) Integrals of an impulse function
d) Sum of successive impulse function
Answer: a
Explanation: All the function derived from an impulse function(successive derivatives and
integrals) are called singularity functions. Here, impulse function is taken as a generalized
function than an ordinary function.

239. What properties does a Continuous time unit Impulse function follow?
a) Shifting, sampling, differentiation, multiplication
b) Multiplication, sampling, shifting
c) Shifting, multiplication, differentiation
d) Sampling only
Answer: a
Explanation: Continuous time impulse functions follows all the properties like shifting, scaling,
sampling or multiplication property, differential.

240. Impulse function is an odd function.

a) True
b) False
Answer: b
Explanation: The Impulse Function is an even function. By scaling property of an Impulse
function we can see, ∂(at)=1/|a|∂(t)
So, substituting, ∂(-t)=1/|-1|∂(t) we get ∂(t), hence, it is an even function. (∂ = del operator).

241. Multiplication of a signal with a Unit Impulse function gives the value of the signal at
which the impulse is located.
a) True
b) False
Answer: a
Explanation: Multiplying the signal by a unit impulse samples the value of the signal at the point
at which the impulse is located. That is x(t)*∂(t)=x(t)|t=0=x(0)∂(t).

242. What is a doublet function?

a) Branch of an impulse function
b) The output of an impulse function
c) The first derivative of an impulse function
d) Any continuous time impulse function has another name that is doublet function
Answer: c
Explanation: The first derivative of d∂(t)/∂(t)=∂’(t) is referred to as a doublet function. The

Department of Electrical and Computer Engineering 57

Wollo University Kombolcha Institute of Technology

derivatives of all orders of the impulse functions are also singularity functions. It is defined as
d∂(t)/dt=∂’(t)=0.

243. What is the area under a doublet function?

a) Unity
b) Negative
c) Zero
d) Positive
Answer: c
Explanation: We can explain by-
Integration -infinity to +infinity x(t)∂’(t)dt= negative of Integration -infinity to +infinity
x’(t)∂’(t)dt=-x’(t)|t=0=-X’(0), where x(t) is any continuous function having a continuous
derivative at t=0. This is ∫∂’(t)=0.

244. How are discrete unit impulse functions and discrete time unit step functions related?
a) They are inverse of each other
b) ∂(n)=u(n)-u(n-1)
c) ∂(n)=u(n)*2∂
d) Integration of unit step function gives unit step function.
Answer: b
Explanation: From definition of u(n) and u(n-1),
u(n) – u(n-1)=∂(n)+sigma k=1 to infinity∂(n-k)- sigma k=1 to infinity ∂(n-k) = ∂(n). In
continuous time, ∂(t)=du(t)/dt.

BIBO Stability & Systems in the Time

Domain
245. What is the full form of BIBO?
a) Boundary input Boundary Output
b) Boundary Input Bounded Output
c) Bonded Input Bonded Output
d) Bounded Input, Bounded Output
Answer: d , Bounded Input, Bounded Output

246. When is a system said to be BIBO stable?

a) When the boundary conditions of the system are stable
b) When there is stability in the overall system
c) Every Bounded input results in a bounded output
d) When the input and output conditions are stable
Answer: c
Explanation: A system is said to be stable if, for any bounded input x(t), the response y(t) is also
Bounded.
i.e |x(t)|≤Bx<∞ implies |y(t)≤Bx<∞.

Department of Electrical and Computer Engineering 58

Wollo University Kombolcha Institute of Technology

247. When does a signal say to be bounded?

a) When it is stable
b) When it gives slow responses
c) Magnitude does not grow without bound
d) When it has small inputs
Answer: c
Explanation: A signal x(t) is said to be bounded if its magnitude does not grow without bound.
i.e |x(t)|≤Bx<∞.

248. How do you describe a stable system informally?

a) When small inputs lead to responses that do not diverge
b) When small inputs lead to responses that diverge
c) When large inputs lead to diverging outputs
d) All inputs lead to outputs that converge
Answer: a
Explanation: When small inputs lead to output responses that do not tend to infinity. The output
of such systems does not diverge if the input does not diverge.

249. The system is stable when y(t)= tx(y).

a) True
b) False
Answer: b
Explanation: Here we can see,
|x(t)|≤Bx<∞, for all t.
Using the input output relation we have y(t)= tx(y). And so we may write,
|Y(t)|=|tx(t)|=|t||x(t)|=|t|Bx.
As tends to infinity, the output also tends to infinity. That is it is unbounded. So it is unstable.

250. How is a time domain system analyzed?

a) Study of a system in accordance to changes in its inputs over time
b) Study of a system in accordance to changes in its over time
c) Study of a system in accordance to changes in its overall structure over time
d) Study of a system in accordance to how a system change itself overall in a time
Answer: d
Explanation: Analysis in the time domain by done in how signals behave over time. That is, a
system or a signal is studied in accordance to how it changes itself overall in time.

251. What is the frequency domain?

a) Analysis of signals in a frequency range
b) Analysis of signals in their bandwidth
c) Analysis of a signal with respect to its frequency
d) Study of a system in accordance to changes in its overall frequency
Answer: c
Explanation: Though this answer is a bit confusing frequency domain is defined as an analysis of
a signal or a system with respect to its frequency. This concept has emerged from the
transformations and the ‘spectrum’ concept.

Department of Electrical and Computer Engineering 59

Wollo University Kombolcha Institute of Technology

252. Time domain is easier for mathematical operation than frequency domain.
a) True
b) False
Answer: b
Explanation: Time domain analysis is much tedious and difficult to perform when it comes to
lengthy solvable problems. Whereas, in a frequency domain, it is very quick to perform. Even
stability is easily attained in frequency domain analysis.

253. What are the mathematical tools to convert a system from a time domain to frequency
domain?
a) Fourier series, Fourier transform, Laplace transform, Z-transform
b) Fourier series only
c) Fourier series and Laplace transform only
d) Fourier series, Fourier transform and Laplace transform only
Answer: a
Explanation: Fourier series, Fourier transform, Laplace transform, z-transform are some tools to
convert a system from a time domain to frequency domain analysis to make it simpler. In fact,
the concept of frequency domain has emerged from these transformations. It was first given by
Joseph Fourier.

254. One of the main limitations of time domain analysis is the noise and frequency.
a) True
b) False
Answer: a
Explanation: True, one of the main limitations of time domain analysis is the noise and
frequency. This is because it is easier in the frequency domain to read it and detect it and solve it.
Time domain analysis is much tedious and difficult to perform when it comes to lengthy solvable
problems.

Continuous Time Convolution

For all the following questions, ‘*’ indicates convolution. $ indicates integral

255. Find the value of h[n]*d[n-1], d[n] being the delta function.
a) h[n-2].
b) h[n].
c) h[n-1].
d) h[n+1].
Answer: c
Explanation: Convolution of a function with a delta function shifts accordingly.

256. Evaluate (exp(-at)u(t))*u(t), u(t) being the heaviside function.

a) (1-exp(at)) u(t)/a
b) (1-exp(at)) u(-t)/a
c) (1-exp(-at)) u(t)/a

Department of Electrical and Computer Engineering 60

Wollo University Kombolcha Institute of Technology

d) (1+exp(-at)) u(t)/a
Answer: c
Explanation: Use the convolution formula.

257. Find the value of h[n]*d[n-5], d[n] being the delta function.
a) h[n-2].
b) h[n-5].
c) h[n-4].
d) h[n+5].
Answer: b
Explanation: Convolution of a function with a delta function shifts accordingly.

258. Evaluate (exp(-4t)u(t))*u(t), u(t) being the heaviside function.

a) (1-exp(4t)) u(t)/a
b) (1-exp(-4t)) u(t)/a
c) (1-exp(=4t)) u(t)/a
d) (1+exp(-4t)) u(t)/a
Answer: b
Explanation: Use the convolution formula.

259. Find the value of h[n-1]*d[n-1], d[n] being the delta function.
a) h[n-2].
b) h[n].
c) h[n-1].
d) h[n+1].
Answer: a
Explanation: Convolution of a function with a delta function shifts accordingly.

260. Find the convolution of x(t) = exp(2t)u(-t), and h(t) = u(t-3)

a) 0.5exp(2t-6) u(-t+3) + 0.5u(t-3)
b) 0.5exp(2t-3) u(-t+3) + 0.8u(t-3)
c) 0.5exp(2t-6) u(-t+3) + 0.5u(t-6)
d) 0.5exp(2t-6) u(-t+3) + 0.8u(t-3)
Answer: a
Explanation: Divide it into 2 sectors and apply the convolution formula.

261. Find the value of h[n]*d[n+1], d[n] being the delta function.
a) h[n-2].
b) h[n].
c) h[n-1].
d) h[n+1].
Answer: d
Explanation: Convolution of a function with a delta function shifts accordingly.

262. Find the convolution of x(t) = exp(3t)u(-t), and h(t) = u(t-3)

a) 0.33exp(2t-6) u(-t+3) + 0.5u(t-3)

Department of Electrical and Computer Engineering 61

Wollo University Kombolcha Institute of Technology

b) 0.5exp(4t-3) u(-t+3) + 0.8u(t-3)

c) 0.33exp(2t-6) u(-t+3) + 0.5u(t-6)
d) 0.33exp(3t-6) u(-t+3) + 0.33u(t-3)
Answer: d
Explanation: Divide it into 2 sectors and apply the convolution formula.

263. Find the value of d(t-34)*x(t+56), d(t) being the delta function.
a) x(t + 56)
b) x(t + 32)
c) x(t + 22)
d) x(t – 22)
Answer: c
Explanation: Convolution of a function with a delta function shifts accordingly.

264. Find x(t)*u(t)

a) tx(t)
b) t2x(t)
c) $x(t2)
d) $x(t)
Answer: d
Explanation: Apply the convolution formula. The above corollary exists for any x(t) [not
impulsive].

LTI Systems – Properties

265. What is the rule h*(x+y) = (y+x)*h called?
a) Commutativity rule
b) Associativity rule
c) Distributive rule
d) Transitive rule
Answer: a
Explanation: By definition, the commutative rule h*x=x*h.

266. Does the system h(t) = exp([-1-2j]t) correspond to a stable system?

a) Yes
b) No
c) Marginally Stable
d) None of the mentioned
Answer: c
Explanation: The system corresponds to an oscillatory system, this resolving to a marginally
stable system

Department of Electrical and Computer Engineering 62

Wollo University Kombolcha Institute of Technology

267. What is the rule h*(x*c) = (x*h)*c called?

a) Commutativity rule
b) Associativity rule
c) Distributive rule
d) Associativity and Commutativity rule
Answer: d
Explanation: By definition, the commutative rule i h*x=x*h and associativity rule = h*(x*c) =
(h*x)*c.

268. Is y[n] = n*cos(n*pi/4)u[n] a stable system?

a) Yes
b) No
c) Marginally stable
d) None of the mentioned
Answer: b
Explanation: The ‘n’ term in the y[n] will dominate as it reaches to infinity, and hence could
reach infinite values.

269. What is the rule (h*x)*c = h*(x*c) called?

a) Commutativity rule
b) Associativity rule
c) Distributive rule
d) Transitive rule
Answer: b
Explanation: By definition, the associativity rule = h*(x*c) = (h*x)*c.

270. Is y[n] = n*sin(n*pi/4)u[-n] a stable system?

a) Yes
b) No
c) Marginally stable
d) None of the mentioned
Answer: b
Explanation: The ‘n’ term in the y[n] will dominate as it reaches to negative infinity, and hence
could reach infinite values. Eventhough + infinity would not be a problem, still the resultant
system would be unstable.

271. What is the following expression equal to: h*(c*(b+d(t))), d(t) is the delta function
a) h*c + h*b
b) h*c*b + b
c) h*c*b + h*c
d) h*c*b + h
Answer: c
Explanation: Apply commutative and associative rules

272. Does the system h(t) = exp([1-4j]t) correspond to a stable system?

a) Yes

Department of Electrical and Computer Engineering 63

Wollo University Kombolcha Institute of Technology

b) No
c) Marginally Stable
d) None of the mentioned
Answer: b
Explanation: The system corresponds to an unstable system, as the Re(exp) term is a positive
quantity.

273. The system transfer function and the input if exchanged will still give the same response.
a) True
b) False
Answer: a
Explanation: By definition, the commutative rule i h*x=x*h=y. Thus, the response will be the
same.

274. For an LTI discrete system to be stable, the square sum of the impulse response should be
a) Integral multiple of 2pi
b) Infinity
c) Finite
d) Zero
Answer: c
Explanation: If the square sum is infinite, the system is an unstable system. If it is zero, it means
h(t) = 0 for all t. However, this cannot be possible. Thus, it has to be finite.

Fourier Series
275. What is Fourier series?
a) The representation of periodic signals in a mathematical manner is called a Fourier series
b) The representation of non periodic signals in a mathematical manner is called a Fourier series
c) The representation of non periodic signals in terms of complex exponentials or sinusoids is
called a Fourier series
d) The representation of periodic signals in terms of complex exponentials or sinusoids is called
a Fourier series
Answer: d
Explanation: The Fourier series is the representation of non periodic signals in terms of complex
exponentials, or equivalently in terms of sine and cosine waveform leads to Fourier series. In
other words, Fourier series is a mathematical tool that allows representation of any periodic wave
as a sum of harmonically related sinusoids.

276. Who discovered Fourier series?

a) Jean Baptiste de Fourier
b) Jean Baptiste Joseph Fourier
c) Fourier Joseph
d) Jean Fourier
Answer: b
Explanation: The Fourier series is the representation of non periodic signals in terms of complex

Department of Electrical and Computer Engineering 64

Wollo University Kombolcha Institute of Technology

exponentials or sine or cosine waveform. This was discovered by Jean Baptiste Joseph Fourier in
18th century.

277. Fourier series representation can be used in case of Non-periodic signals too. True or false?
a) True
b) False
Answer: b

Explanation: False. The Fourier series is the representation of periodic signals in terms of
complex exponentials, or equivalently in terms of sine and cosine waveform leads to Fourier
series. In other words, Fourier series is a mathematical tool that allows representation of any
periodic wave as a sum of harmonically related sinusoids. They are for periodic signals only.

278. What are the conditions called which are required for a signal to fulfil to be represented as
Fourier series?
a) Dirichlet’s conditions
b) Gibbs phenomenon
c) Fourier conditions
d) Fourier phenomenon
Answer: a
Explanation: When the Dirichlet’s conditions are satisfied, then only for a signal, the fourier
series exist. Fourier series is of two types- trigonometric series and exponential series.

279. Choose the condition from below that is not a part of Dirichlet’s conditions?
a) If it is continuous then there are a finite number of discontinuities in the period T1
b) It has a finite average value over the period T
c) It has a finite number of positive and negative maxima in the period T
d) It is a periodic signal
Answer: d
Explanation: Even if the Fourier series demands periodicity as the major necessity for its
formation still it is not a part of Dirichlet’s condition. It is the basic necessity for Fourier series.

280. What are the two types of Fourier series?

a) Trigonometric and exponential
b) Trigonometric and logarithmic
c) Exponential and logarithmic
d) Trigonometric only
Answer: a
Explanation: The two types of Fourier series are- Trigonometric and exponential. The
exponential is more convenient for Fourier series calculations.

281. How is a trigonometric Fourier series represented?

a) A0 +∑[ancos(w0t)+ ansin(w0t)]
b) ∑[ancos(w0t)+ ansin(w0t)]
c) A0 *∑[ancos(w0t)+ ansin(w0t)]
d) A0 +∑[ancos(w0t)+ ansin(w0t)] + sinwt

Department of Electrical and Computer Engineering 65

Wollo University Kombolcha Institute of Technology

Answer: a
Explanation: A0 + ∑[ancos(w0t)+ ansin(w0t)] is the correct representation of a trigonometric
Fourier series. Here A0 = 1/T∫x(t)dt and an =2/T∫x(t)cos(w0t)dt and bn= 2/T∫x(t)sin(w0t)dt.

282. How is the exponential Fourier series represented?

a) X(t) = ∑Xnejnwt + wt
b) X(t) = 1/T∑Xnejnwt
c) X(t) = ∑Xnejnwt
d) X(t) = T*∑Xnejnwt
Answer: c
Explanation: The exponential Fourier series is represented as – X(t)=∑Xnejnwt. Here, the X(t) is
the signal and Xn=1/T∫x(t)e-jnwt.

283. What is the equation – X(t)=∑Xnejnwt called?

a) Synthesis equation
b) Analysis equation
c) Frequency domain equation
d) Discrete equation
Answer: a
Explanation: The equation – X(t) = ∑Xnejnwt called the synthesis equation of an exponential
Fourier series. It is because it is used to synthesize the Fourier series.

284. What is the equation – Xn=1/T∫x(t) ejwtn called?

a) Synthesis equation
b) Analysis equation
c) Frequency domain equation
d) Discrete equation
Answer: b
Explanation: The equation – Xn=1/T∫x(t)e-jwtn called the analysis equation of an exponential
Fourier series. It is because it is used to synthesize the Fourier series

Fourier Series & Coefficients

285. What are fourier coefficients?
a) The terms that are present in a fourier series
b) The terms that are obtained through fourier series
c) The terms which consist of the fourier series along with their sine or cosine values
d) The terms which are of resemblance to fourier transform in a fourier series are called fourier
series coefficients
Answer: c
Explanation: The terms which consist of the fourier series along with their sine or cosine values
are called fourier coefficients. Fourier coefficients are present in both exponential and
trigonometric fourier series.

Department of Electrical and Computer Engineering 66

Wollo University Kombolcha Institute of Technology

286. Which are the fourier coefficients in the following?

a) a0, an and bn
b) an
c) bn
d) an and bn
Answer: a
Explanation: These are the fourier coefficients in a trigonometric fourier series.
a0 = 1/T∫x(t)dt
an = 2/T∫x(t)cos(nwt)dt
bn = 2/T∫x(t)sin(nwt)dt

287. Do exponential fourier series also have fourier coefficients to be evaluated.

a) True
b) False
Answer: a
Explanation: The fourier coefficient is : Xn = 1/T∫x(t)e-njwtdt.

288. The fourier series coefficients of the signal are carried from –T/2 to T/2.
a) True
b) False
Answer: a
Explanation: Yes, the coefficients evaluation can be done from –T/2 to T/2. It is done for the
simplification of the signal.

289. What is the polar form of the fourier series?

a) x(t) = c0 + ∑cncos(nwt+ϕn)
b) x(t) = c0 + ∑cncos(ϕn)
c) x(t) = ∑cncos(nwt+ϕn)
d) x(t) = c0+ ∑cos(nwt+ϕn)
Answer: a
Explanation: x(t) = c0 + ∑cncos(nwt+ϕn), is the polar form of the fourier series.
C0=a0 and cn = √a2n+ b2n for n≥1
And ϕn = tan-1 bn/an .

290. What is a line spectrum?

a) Plot showing magnitudes of waveforms are called line spectrum
b) Plot showing each of harmonic amplitudes in the wave is called line spectrum
c) Plot showing each of harmonic amplitudes in the wave is called line spectrum
d) Plot showing each of harmonic amplitudes called line spectrum
Answer: b
Explanation: The plot showing each of harmonic amplitudes in the wave is called line spectrum.
The line rapidly decreases for waves with rapidly convergent series.

291. Fourier series is not true in case of discrete time signals.

a) True
b) False

Department of Electrical and Computer Engineering 67

Wollo University Kombolcha Institute of Technology

Answer: b
Explanation: Fourier series is also true in case of discrete time signals. They just need to follow
the dirichlet’s conditions.

292. What is the disadvantage of exponential Fourier series?

a) It is tough to calculate
b) It is not easily visualized
c) It cannot be easily visualized as sinusoids
d) It is hard for manipulation
Answer: c
Explanation: The major disadvantage of exponential Fourier series is that it cannot be easily
visualized as sinusoids. Moreover, it is easier to calculate and easy for manipulation leave aside
the disadvantage.

293. Fourier series uses which domain representation of signals?

a) Time domain representation
b) Frequency domain representation
c) Both combined
d) Neither depends on the situation
Answer: b
Explanation: Fourier series uses frequency domain representation of signals. X(t)=1/T∑Xnejnwt.
Here, the X(t) is the signal and Xn = 1/T∫x(t)e-jwtn.

294. How does Fourier series make it easier to represent periodic signals?
a) Harmonically related
b) Periodically related
c) Sinusoidally related
d) Exponentially related
Answer: a
Explanation: Fourier series makes it easier to represent periodic signals as it is a mathematical
tool that allows the representation of any periodic signals as the sum of harmonically related
sinusoids

Fourier Series Properties

295. How do we represent a pairing of a periodic signal with its fourier series coefficients in case
of continuous time fourier series?
a) x(t) ↔ Xn
b) x(t) ↔ Xn+1
c) x(t) ↔ X
d) x(n) ↔ Xn
Answer: a
Explanation: In case of continuous time fourier series, for simplicity, we represent a pairing of a
periodic signal with its fourier series coefficients as,

Department of Electrical and Computer Engineering 68

Wollo University Kombolcha Institute of Technology

x(t) ↔ Xn
here, x(t) is the signal and Xn is the fourier series coefficient.

296. What are the properties of continuous time fourier series?

a) Linearity, time shifting
b) Linearity, time shifting, frequency shifting
c) Linearity, time shifting, frequency shifting, time reversal, time scaling, periodic convolution
d) Linearity, time shifting, frequency shifting, time reversal, time scaling, periodic convolution,
multiplication, differentiation
Answer: d
Explanation: Linearity, time shifting, frequency shifting, time reversal, time scaling, periodic
convolution, multiplication, differentiation are some of the properties followed by continuous
time fourier series. Integration and conjugation are also followed by continuous time fourier
series.

297. Integration and conjugation are also followed by continuous time fourier series?
a) True
b) False
Answer: a
Explanation: Linearity, time shifting, frequency shifting, time reversal, time scaling, periodic
convolution, multiplication, differentiation are some of the properties followed by continuous
time fourier series. Integration and conjugation are also followed by continuous time fourier
series.

298. If x(t) and y(t) are two periodic signals with coefficients Xn and Yn then the linearity is
represented as?
a) ax(t) + by(t) = aXn + bYn
b) ax (t) + by(t) = Xn + bYn
c) ax(t) + by(t) = aXn + Yn
d) ax(t) + by(t) = Xn + Yn
Answer: a
Explanation: ax(t) + by(t) = aXn + bYn, x(t) and y(t) are two periodic signals with coefficients Xn
and Yn.

299. How is time shifting represented in case of periodic signal?

a) If x(t) is shifted to t0, Xn is shifted to t0
b) x(t-t0), Yn = Xn e-njwt0
c) Xn = x(t-t0), Yn = Xn e-njwt0
d) Xn = x(-t0), Yn = Xn e-njwt0
Answer: c
Explanation: If x(t) and y(t) are two periodic signals with coefficients Xn and Yn, then if a signal
is shifted to t0, then the property says,
Xn = x(t-t0), Yn = Xne-njwt0

Department of Electrical and Computer Engineering 69

Wollo University Kombolcha Institute of Technology

300. What is the frequency shifting property of continuous time fourier series?
a) Multiplication in the time domain by a real sinusoid
b) Multiplication in the time domain by a complex sinusoid
c) Multiplication in the time domain by a sinusoid
d) Addition in the time domain by a complex sinusoid
Answer: b
Explanation: If x(t) and y(t) are two periodic signals with coefficients Xn and Yn,
Then y(t)= ejmwtx(t)↔Yn=Xn-m.
Hence, we can see that a frequency shift corresponds to multiplication in the time domain by
complex sinusoid whose frequency is equal to the time shift.

301. What is the time reversal property of fourier series coefficients?

a) Time reversal of the corresponding sequence of fourier series
b) Time reversal of the last term of fourier series
c) Time reversal of the corresponding term of fourier series
d) Time reversal of the corresponding sequence
Answer: a
Explanation: x(t)↔ Xn
Y(t) = x(-t)↔Yn=X-n.
That is the time reversal property of fourier series coefficients is time reversal of the
corresponding sequence of fourier series.

302. It does not depend whether the signal is odd or even, it is always reversal of the
corresponding sequence of fourier series.
a) True
b) False
Answer: b
Explanation: It does depend whether the signal is odd or even.
If the signal is even, the reversal is positive and if the signal is odd, the reversal is negative.

303. Why does the signal change while time scaling?

a) Because the frequency changes
b) Time changes
c) Length changes
d) Both frequency and time changes
Answer: a
Explanation: x(t)↔Xn
Y(t) = x(at)↔Yn = Xn
Hence, the fourier coefficients have not changed but the representation has changed because of
changes in fundamental frequency.

304. What is the period of the signal when it is time shifted?

a) Changes according to the situation
b) Different in different situation

Department of Electrical and Computer Engineering 70

Wollo University Kombolcha Institute of Technology

c) Remains the same

d) Takes the shifted value
Answer: c
Explanation: The period of the periodic signal does not change even if it is time shifted.
If x(t) and y(t) are two periodic signals with coefficients Xn and Yn, then if a signal is shifted to
t0, then the property says,
Xn = x(t-t0), Yn = Xne-njwt0.

Exponential Fourier Series and Fourier

Transforms
305. The Fourier transform of u (t) is B (jω) and the Laplace transform of u (t) is A(s). Which of
the following is correct?
a) B(jω) = A(s)
b) A(s) = 1s but B(jω) ≠ 1jω
c) A(s) ≠ 1s but B(jω) ≠ 1jω
d) A(s) ≠ 1s but B(jω) = 1jω
Answer: b
Explanation: Laplace transform of u (t) is given by u (t) –> A(s) = 1s
Fourier transform of u (t) is given by, u (t) = B (jω) = (1jω) + π δ (ω)
Therefore A(s) = 1s but B(jω) ≠ 1jω is satisfied.

306. Given, X (ejω) = (b−a)ejωe−j2ω−(a+b)ejω+ab, |b|<1<|a|

The value of x[n] is __________
a) bn u [n] + an u [n-1]
b) bn u [n] – an u [-n-1]
c) bn u [n] + an u [-n-1]
d) bn u [n] – an u [n+1]
Answer: c
Explanation: X (ejω) = (b−a)ejωe−j2ω−(a+b)ejω+ab
= (b−a)ejω1−(a+b)e−jω+abe−j2ω
= 11−be−jω+(−1)1−ae−jω
∴ x [n] = bn u [n] + an u [-n-1].

307. The input and output of an LTI system are x (t) = e-3t u (t) and y (t) = e-t u (t). The
differential equation which characterizes the system is ___________
a) dy(t)dt+y(t)=dx(t)dt+3x(t)
b) dy(t)dt+2y(t)=dx(t)dt+3x(t)
c) dy(t)dt–y(t)=dx(t)dt+3x(t)
d) dy(t)dt–2y(t)=dx(t)dt+3x(t)
Answer: a
Explanation: X (s) = 1s+3, Y (s) = 1s+1

Department of Electrical and Computer Engineering 71

Wollo University Kombolcha Institute of Technology

∴ H(s) = Y(s)X(s)=1/(s+1)1/(s+3)=s+3s+1
Now, s Y(s) + Y(s) = s X(s) + 3 X(s)
So, the differential equation together with the condition of initial rest that characterizes the
system is dy(t)dt+y(t)=dx(t)dt+3x(t).

308. The Fourier transform of signal e-2t u(t-3) is ___________

a) e−3(2−jω)2−jω
b) e−3(2+jω)2+jω
c) e3(2−jω)2−jω
d) e3(2+jω)2+jω
Answer: b
Explanation: X (jω) = ∫∞−∞x(t)e−jωtdt
Given u (t-3), hence value of this will be equal to 1 when t>=3
∴ X (jω) = ∫∞3e−2te−jωtdt
= e−3(2+jω)2+jω.

309. The Fourier transform of the signal e-4|t| is ____________

a) 816+ω2
b) −816+ω2
c) 416+ω2
d) −416+ω2
Answer: a
Explanation: X (jω) = ∫∞−∞e−4|t|e−jωtdt
Now, e-4|t| = e-4t, when t>0 and
e4t, when t<0
∴X (jω) = ∫0−∞e4te−jωtdt+∫∞0e−4te−jωtdt
= 816+ω2.

310. The Inverse Fourier transform of the signal e-2|ω| is ____________

a) 2π(4+t2)
b) 12π(4+t2)
c) 1π(4+t2)
d) 1(4+t2)
Answer: a
Explanation: e-2|ω| = e-2ω, ω > 0 and e2ω, ω<0
Hence, x (t) = 12π∫∞−∞e−2|ω|e−jωtdω
= 12π∫0−∞e2ωe−jωtdω+12π∫∞0e−2ωe−jωtdω
= 2π(4+t2).

311. The inverse Laplace transform of F(s) = 2s+ce−bs is _____________

a) 2e-k (t+b) u (t+b)
b) 2e-k (t-b) u (t-b)
c) 2ek (t-b) u (t-b)

Department of Electrical and Computer Engineering 72

Wollo University Kombolcha Institute of Technology

d) 2ek (t-b) u (t+b)

Answer: b
Explanation: Let G(s) = 2s+c
Or, G (t) = L-1{G(s)} = 2e-ct
∴ F (t) = L-1{G(s) e-bs}
= 2e-k (t-b) u (t-b).

312. The Laplace transform of the function e-2tcos(3t) + 5e-2tsin(3t) is ____________

a) (s+2)−15(s+2)2−9
b) (s+2)+15(s+2)2−9
c) (s+2)+15(s+2)2+9
d) (s+2)−15(s+2)2+9
Answer: c
Explanation: L {e-2tcos(3t) + 5e-2tsin(3t)}
= (s+2)(s+2)2+9+53(s+2)2+9
= (s+2)+15(s+2)2+9.

313. A band pass signal extends from 1 KHz to 2 KHz. The minimum sampling frequency that is
needed to retain all information of the sampled signal is ___________
a) 1 KHz
b) 2 KHz
c) 3 KHz
d) 4 KHz
Answer: b
Explanation: We know that the minimum sampling frequency is twice the maximum bandwidth.
Here, maximum bandwidth = 2-1 = 1 KHz
So, Minimum sampling frequency = 2(Bandwidth) = 2(2-1) = 2 KHz.

314. The Laplace transform of the function 6e5tcos(2t) – e7t is ______________

a) 6(s−5)(s−5)2+4–1s−7
b) 6(s−5)(s−5)2+4+1s−7
c) 6(s+5)(s+5)2+4–1s−7
d) 6(s+5)(s+5)2+4+1s−7
Answer: a
Explanation: We know that, Laplace transform of eat = 1s−a
Here, a=7, so L {e7t} = 1s−7
And the Laplace transform of eat cos (b t) = (s−a)(s−a)2+b2
Here, a=5 and b=2, so L {6e5t cos (2t)} = 6(s−5)(s−5)2+4
∴ L {6e5tcos(2t) – e7t} = 6(s−5)(s−5)2+4–1s−7.

315. The inverse Laplace transform of F(s) = e−3ss(s2+3s+2) is ______________

a) {0.5 + 0.5e-2(t+3)-e-(t+3)} u (t+3)
b) {0.5 + 0.5e-2(t-3)-e-(t-3)} u (t-3)
c) {0.5 – 0.5e-2(t-3)-e-(t-3)} u (t-3)

Department of Electrical and Computer Engineering 73

Wollo University Kombolcha Institute of Technology

d) 0.5 + 0.5e-2t-e-t)
Answer: b
Explanation: Let G(s) = 1s(s2+3s+2)
Or, F(s) = G(s) e-3s
G (t) = L-1{G(s)}
= L-1{As+Bs+2+Cs+1}
Solving we get, A=0.5, B=0.5, C=-1
So, G (t) = 0.5 + 0.5e-2t-e-t
The inverse Laplace transform is F (t) = {0.5 + 0.5e-2(t-3)-e-(t-3)} u (t-3).

316. The Fourier transform of the signal e-t+2 u (t-2) is ___________

a) e−2jω1−2ω
b) e2jω1+2ω
c) e−2jω1+2ω
d) e2jω1−2ω
Answer: c
Explanation: Fourier transform of e-t u(t) = 11+jω
∴ The Fourier transform of x (t-2) = e-2jω X (jω) [Using Shifting property]
Hence, X (jω) = e−2jω1+2ω.

Fourier Transforms
317. Which of the following is the Analysis equation of Fourier Transform?
a) F(ω)=∫∞−∞f(t)ejωtdt
b) F(ω)=∫∞0f(t)e−jωtdt
c) F(ω)=∫∞0f(t)ejωtdt
d) F(ω)=∫∞−∞f(t)e−jωtdt
Answer: d
Explanation: For converting time domain to frequency domain, we use analysis equation. The
Analysis equation of Fourier Transform is F(ω)=∫∞−∞f(t)e−jωtdt.

318. Choose the correct synthesis equation.

a) f(t)=12π∫∞−∞F(ω)e−jωtdω
b) f(t)=12π∫∞−∞F(ω)ejωtdω
c) f(t)=12π∫∞0F(ω)e−jωtdω
d) f(t)=12π∫∞0F(ω)ejωtdω
Answer: b
Explanation: Synthesis equation converts from frequency domain to time domain. The synthesis
equation of fourier transform is f(t)=12π∫∞−∞F(ω)ejωtdω.

319. Find the fourier transform of an exponential signal f(t) = e-at u(t), a>0.
a) 1a+jω
Department of Electrical and Computer Engineering 74
Wollo University Kombolcha Institute of Technology

b) 1a−jω
c) 1−a+jω
d) 1−a−jω
Answer: a
Explanation: Given f(t)= e-at u(t)
We know that u(t)={01t<0t>0
Fourier transform,
F(ω)=∫∞−∞f(t)e−jωtdt=∫∞−∞e−atu(t)e−jωtdt=∫∞0e−(a+jω)tdt
F(ω) = 1a+jω, a>0.

320. Find the fourier transform of the function f(t) = e-a|t|, a>0.
a) 2aa2−ω2
b) 2aa2+ω2
c) 2aω2−a2
d) aa2+ω2
Answer: b
Explanation: The given two-sided exponential function f(t) = e-a|t|, a>0 can be expressed as
f(t)={e−ateatt≥0t≤0
The Fourier transform is
F(ω)=∫∞−∞f(t)e−jωtdt=∫0−∞f(t)e−jωtdt+∫∞0f(t)e−jωtdt
F(ω)=1a+jω+1a−jω=2aa2+ω2.
312. Gate function is defined as ______________
a) G(t)={10|t|<τ2elsewhere
b) G(t)={10|t|>τ2elsewhere
c) G(t)={10|t|≤τ2elsewhere
d) G(t)={10|t|≥τ2elsewhere
Answer: a
Explanation: A gate function is a rectangular function defined as
G(t)=rect(tτ)={10|t|<τ2elsewhere
Where τ is pulse width.

322. Find the fourier transform of the gate function.

a) 1ωsin(ωτ2)
b) 1ωcos(ωτ2)
c) 2ωsin(ωτ2)
d) 2ωcos(ωτ2)
Answer: c
Explanation: Gate function is defined as
G(t)={10|t|<τ2elsewhere
The fourier transform is F(ω)=∫∞−∞f(t)e−jωtdt=∫τ/2−τ/2e−jωtdt=2ωsin(ωτ2).

Department of Electrical and Computer Engineering 75

Wollo University Kombolcha Institute of Technology

323. Choose the wrong option.

a) G(t) = rect(tτ)
b) G(t) = u(t + τ2) – u(t-τ2)
c) G(ω) = τ sa(wτ2)
d) G(f) = τ sinc(f)
Answer: d
Explanation: Fourier transform of gate function, G(ω) = 2ωsin(wτ2)
Multiplying and dividing by τ we get
G(ω)=τsin(wτ2)wτ2=τsin(2πfτ2)2πfτ2=τsin(πτf)πτf=τsinc(τf).
324. Bandwidth of the gate function is __________
a) τ Hz
b) 1τ Hz
c) 2τ Hz
d) 2τ Hz
Answer: b
Explanation: The practical bandwidth of the gate function corresponds to the first zero crossing
in the spectrum. Therefore, the bandwidth of the pulse or gate function is 2πτ or 1τ Hz.

325. Which of the following is not a fourier transform pair?

a) u(t)↔πδ(ω)+1jω
b) sgn(t)↔2jω
c) A↔2πδ(ω2)
d) G(t)↔sa(ωτ2)
Answer: d
Explanation: G(t)↔sa(ωτ2) is not a fourier transform pair.
G(t)↔τsa(ωτ2) (or) G(t)↔G(t)τsinc(τf).
326. Find the fourier transform of the unit step function.
a) πδ(ω) + 1ω
b) πδ(ω) + 1jω
c) πδ(ω) – 1jω
d) δ(ω) + 1jω
Answer: b
Explanation: We know that sgn(t) = 2u(t) – 1.
u(t) = 12[sgn(t)+1] Its Fourier transform is F[u(t)] = 12 F[sgn(t)] + 12 F[1]
As the Fourier transforms F[1] = 2πδ(ω) and [sgn(t)] = 2jω, hence
F[u(t)] = πδ(ω) + 1jω

Inverse Fourier Transform

327. Find the inverse Fourier transform of X(ω) = e-2ω u(ω).
a) 12π(2+jt)
Department of Electrical and Computer Engineering 76
Wollo University Kombolcha Institute of Technology

b) 12π(2−jt)
c) 12(2+jt)
d) 1π(2+jt)
Answer: b
Explanation: We know that x(t) = 12π∫∞−∞X(ω)ejωtdω
x(t) = 12π∫∞−∞e−2ωu(ω)ejωtdω=12π∫∞−∞e−2ωejωtdω=12π(2−jt).

328. Find the inverse Fourier transform of X(ω) = 1+3(jω)(3+jω)2.

a) 3e-3t u(t) + 8e-3t u(t)
b) 3te-3t u(t) – 8e-8t u(t)
c) 3e-3t u(t) + 8te8t u(t)
d) 3e-3t u(t) – 8te-3t u(t)
Answer: d
Explanation: Given X(ω) = 1+3(jω)(3+jω)2=A3+jω+B(3+jω)2=33+jω–8(3+jω)2
Applying inverse Fourier transform, we get
x(t) = 3e-3t u(t) – 8te-3t u(t).

329. Find the inverse Fourier transform of δ(ω).

a) 12π
b) 2π
c) 1π
d) π
Answer: d
Explanation: We know that x(t) = 12π∫∞−∞X(ω)ejωtdω
= 12π∫∞−∞δ(ω)ejωtdω=12π.

330. Find the inverse Fourier transform of u(ω).

a) 12δ(t)+j2πt
b) 12δ(t)–j2πt
c) δ(t) + j2πt
d) δ(t) – j2πt
Answer: a
Explanation: We know that u(ω) = 12[1+sgn(ω)].
Applying linearity property,
u(ω) = -1 [12]+F−1[12sgn(ω)]
u(ω) = 12δ(t)+j2πt.

331. Find the inverse Fourier transform of ej2t.

a) 2πδ(ω-2)
b) πδ(ω-2)
c) πδ(ω+2)
d) 2πδ(ω+2)
Answer: a

Department of Electrical and Computer Engineering 77

Wollo University Kombolcha Institute of Technology

Explanation: We know that ejω0 t ↔ 2πδ(ω-ω0)

∴ ej2t ↔ 2πδ(ω-2).

332. Find the inverse Fourier transform of jω.

a) δ(t)
b) ddt δ(t)
c) 1δ(t)
d) ∫δ(t)
Answer: b
Explanation: Time differentiation property, ddt x(t) ↔ jωX(ω) and we know that δ(t) ↔ 1
∴ ddt δ(t) ↔ jω.

333. Find the inverse Fourier transform of X(ω)=6+4(jω)(jω)2+6(jω)+8.

a) e-2t u(t) – 5e-4t u(t)
b) e-2t u(t) + 5e-4t u(t)
c) -e-2t u(t) – 5e-4t u(t)
d) -e-2t u(t) + 5e-4t u(t)
Answer: d
Explanation: X(ω)=6+4(jω)(jω)2+6(jω)+8=Ajω+2+Bjω+4=−1jω+2+5jω+4
Applying inverse Fourier transform, we get
x(t) = -e-2t u(t) + 5e-4t u(t).

334. Find the convolution of the signals x1 (t) = e-2t u(t) and x2 (t) = e-3t u(t).
a) e-2t u(t) – e-3t u(t)
b) e-2t u(t) + e-3t u(t)
c) e2t u(t) – e3t u(t)
d) e2t u(t) – e-3t u(t)
Answer: a
Explanation: Convolution property, x1 (t)*x2 (t) ↔ X1 (ω) X2 (ω)
∴ x1 (t)*x2 (t) = F-1 [X1 (ω) X2 (ω)]
Given x1 (t) = e-2t u(t)
∴ X1 (ω) = 1jω+2
Given x2 (t) = e-3t u(t)
∴ X1 (ω) = 1jω+3
x1 (t)*x2 (t) = F-1 [X1 (ω) X2 (ω)] = F-1 [1jω+21jω+3]=F−1[1jω+2–1jω+3]
∴ x1 (t)*x2 (t) = e-2t u(t)-e-3t u(t).

335. Find the inverse Fourier transform of f(t)=1.

a) u(t)
b) δ(t)
c) e-t
d) 1jω
Answer: b
Explanation: We know that the Fourier transform of f(t) = 1 is F(ω) = 2πδ(ω).
Replacing ω with t

Department of Electrical and Computer Engineering 78

Wollo University Kombolcha Institute of Technology

F(t) = 2πδ(t)
As per duality property F(t) ↔ 2πf(-ω), we have
2πδ(t) ↔ 2π(1)
δ(t) ↔ 1
Hence, the inverse Fourier transform of 1 is δ(t).

336. Find the inverse Fourier transform of sgn(ω).

a) 1πt
b) jπt
c) jt
d) 1t
Answer: b
Explanation: Given the function F(ω)=sgn(ω). The Fourier transform of a Signum function is
sgn(ω) = 2jω.
Applying the duality property F(t) ↔ 2πf(-ω), we get
F(2jt) = 2πsgn(-ω).
As sgn(ω) is an odd function, sgn(-ω)=-sgn(ω).
Hence, 2jt ↔ -sgn(ω)
Or 2πt ↔ sgn(ω)
Therefore, the inverse Fourier transform of sgn(ω) is jπt

Discrete Fourier Transform

337.Given that S1 and S2 are two discrete time systems. Consider the following statements:
i) If S1 and S2 are linear, then S is linear
ii) If S1 and S2 are non-linear, then S is non-linear
iii) If S1 and S2 are causal, then S is causal
iv) If S1 and S2 are time invariant, then S is time invariant
The true statements from the above are ____________
a) i, ii, iii
b) ii, iii, iv
c) I, iii, iv
d) I, ii, iii, iv
Answer: c
Explanation: Only statement ii is false.
For example, S1: y[n] = x[n] +b and S2: y[n] = x[n]-b, where, b≠0.
S{x[n]} = S2 {S1{x[n]}} = S2{x[n] +b} = x[n]
Hence, S is linear.

338. For two discrete time systems, consider the following statements:
i) If S1 and S2 are linear and time invariant, then interchanging their order does not change the
system.
ii) If S1 and S2 are linear and time variant, then interchanging their order does not change the
system
The correct statement from the above is __________

Department of Electrical and Computer Engineering 79

Wollo University Kombolcha Institute of Technology

a) Both i & ii
b) Only i
c) Only ii
d) Neither i, nor ii
Answer: b
Explanation: S1: y[n] = n x[n]
And S2: y[n] = n x [n+1]
If x[n] = δ[n], then S2 {S1 {δ[n]}} = S2 [0] = 0
S1 {S2 {δ[n]}} = S1 {δ[n+1]} = – δ[n+1]≠0.

339. The following input-output pairs have been observed during the operation of a time
invariant system
i) x1[n] = {1, 0, 2} (Laplace transform) y1 [n] = {0, 1,2}
ii) x2[n] = {0,0, 3} (Laplace transform) y2[n] = {0,1,0,2}
iii) x3[n] = {0,0,0,1} (Laplace transform) y3[n] = {1,2,1}
The conclusion regarding the linearity of the system is _____________
a) Linear
b) Non-linear
c) One more observation is required
d) Conclusion cannot be drawn from observation
Answer: b
Explanation: System is not linear. This is evident from the observation of the pairs, x3[n] – y3[n]
and x2[n] and y2[n]. If the system were linear y2[n] would be of the form y2[n] = {3, 6, 3}.

340. S1 and S2 are two DT systems which are connected together to form a new system. Consider
the following statements:
i) If S1 and S2 are non-causal, then S is non-causal
ii) If S1 and/or S2 are unstable, then S is unstable
The correct statement from the above is ____________
a) Both i and ii
b) Only i
c) Only ii
d) Neither i nor ii
Answer: d
Explanation: S1: y[n] = x [n+1] …… (Non-causal)
S2: y[n] = x [n-2] ……… (Causal)
S: y[n] = x [n-1] which is causal ………. (False)
S1: y[n] = ex[n] stable, S2: y[n] = ln(x[n]) ……… (Unstable)
But S: y[n] = x[n] ………. (Stable, false)

341. Given a signal x[n] = δ[n] + 0.9 δ [n − 6]. The Discrete Time Fourier Transform for 8 points
is __________
a) 1 – 0.9 e−j2π8k6
b) 1 + 0.9 e−j2π8k6
c) 1 + 0.9 ej2π8k6

Department of Electrical and Computer Engineering 80

Wollo University Kombolcha Institute of Technology

d) 1 – 0.9 ej2π8k6
Answer: b
Explanation: Given N = 8.
Now, x[k] = ∑N−10x[n]e−j2πNkn
= ∑70x[n]e−j2π8kn
= ∑7N=0(δ[n]+0.9δ[n–6])e−j2π8kn
= 1 + 0.9 e−j2π8kn
Here, n = 6, from given question.
Hence, x[k] = 1 + 0.9 e−j2π8k6.

342. The Z transform of δ (n − m) is ___________

a) z-n
b) z-m
c) 1z−n
d) 1z−m
Answer: b
Explanation: δ (n − m) is a delayed impulse function which is delayed by m units. We know that
the Z-transform of a delayed function f (n-m) is (z-m) times the Z-transform of the function f (n).
So, the Z transform of δ (n − m) is z-m.

343. A 10 V is connected across a load whose V-I characteristics is given by 7I = V2 + 2V. The
internal resistance of the battery is of magnitude 1Ω. The current delivered by the battery is
____________
a) 6 A
b) 5 A
c) 7 A
d) 8 A
Answer: b
Explanation: 7I = V2 + 2V …………………. (1)
Now, V = 10 – 1 × I
Putting the value of V in eqt (1), we get,
&I = (10 – I) 2 + 2(10 – I) …………………. (2)
Or, I = 100 + I2 – 20I + 20 – 2I
Or, I2 – 29I + 120 = 0
∴ I=+29±292–4(120)√2=29±192
I = 5 A, 24 A
Now, I = 24 A is not possible because V will be negative from eqt (2)
∴ I = 5 A.

344. The period of the signal x(t) = 10 sin 12 π t + 4 cos18 π t is ____________

a) π4
b) 16
c) 19
d) 13

Department of Electrical and Computer Engineering 81

Wollo University Kombolcha Institute of Technology

Answer: d
Explanation: There are two waveforms of frequencies 6 and 9, respectively. Hence the combined
frequency is the highest common factor between 6 and 9 which is 3. Therefore the period is 13.

345. Given a series RLC circuit with V = 5V, R = 200 kΩ, C = 10µF. Sampling frequency of the
circuit is 10 Hz. The expression and the ROC of the z-transform of the sampled signal are
____________
a) 5zz−e−5', |z|<e-5
b) 5zz−e−0.05', |z|<e-0.05
c) 5zz−e−0.05', |z|>e-0.05
d) 5zz−e−5', |z|>e-5
Answer: c
Explanation: I (t) = VRe−t/RC
Voltage across resistor = R I (t)
= V e-t/RC = 5 e-t/RC
= 5 e−t200×10×10−6×103 = 5 e-t/2
Given that, the Sampling frequency of the circuit = 10 Hz
Hence, x (n) = 5e-n/2 X 10 = 5e-0.05n
Now, X (z) = ∑∞n=−∞x[n]z−n
= 5 ∑∞n=−∞(e−0.05z−n)n
= 5. 11−e−0.05Z−1', ROC |z|>e-0.05
= 5zz−e−0.05', |z|>e-0.05.

346. Given a series RLC circuit with V = 5V, R = 200 kΩ, C = 10µF. Sampling frequency of the
circuit is 10 Hz. The samples x (n), where n=0,1,2,…., is ___________
a) 5(1-e-0.05n)
b) 5e-0.05n
c) 5(1-e-5n)
d) 5e-5n
Answer: b
Explanation: The charging current in circuit I (t) = I (0+) e-t/RC
Since the capacitor acts as short circuit, I (0+) = VR
∴ I (t) = VR e-t/RC
Voltage across resistor = R I (t)
= V e-t/RC = 5 e-t/RC
= 5 e−t200×10×10−6×103 = 5 e-t/2
Given that, the Sampling frequency of the circuit is 10 Hz
∴ x (n) = 5e-n/2 X 10 = 5e-0.05n.

347. For the circuit given below, if the frequency of the source is 50 Hz, then a value of to which
results in a transient free response is _________________

Department of Electrical and Computer Engineering 82

Wollo University Kombolcha Institute of Technology

a) 0
b) 1.78 ms
c) 7.23 ms
d) 9.21 ms
Answer: b
Explanation: T = LR
Or, T = 0.015 = 0.002 s = 2 ms
For the ideal case, transient response will die out with time constant.
Practically, T will be less than 2 ms.

348. If G(f) represents the Fourier Transform of a signal g (t) which is real and odd symmetric in
time, then G (f) is ____________
a) Complex
b) Imaginary
c) Real
d) Real and non- negative
Answer: b
Explanation: Fourier transform of g (t) is G (f)
Given that, g (t) is real, odd and symmetric with respect to time.
∴G*(jm) = – G(jm); G(jm) purely imaginary.

349. If R1 is the region of convergence of x (n) and R2 is the region of convergence of y(n), then
the region of convergence of x (n) convoluted y (n) is ___________
a) R1 + R2
b) R1 – R2
c) R1 ∩ R2
d) R1 ∪ R2
Answer: c
Explanation: The z-transform of x (n) = X (z). Let the region of convergence be R1
The z-transform of y (n) = y (z). Let the region of convergence be R2
The z-transform of x (n) * y (n) is X (z).Y (z) [from property]
So, the region of convergence is R1 ∩ R2.

350. The system under consideration is an RC low-pass filter with R = 1 kΩ and C = 1 µF. Let H
(f) denotes the frequency response of the RC, low-pass filter. Let f1 be the highest frequency,
such that 0≤|f|≤f1, |H(f1)|H(0)≥0.95 Then f1 is ___________

Department of Electrical and Computer Engineering 83

Wollo University Kombolcha Institute of Technology

a) 327.8
b) 163.9
c) 52.2
d) 104.4
Answer: c
Explanation: H (ω) = 1jωCR+(1jωC)=11+jωRC
H (f) = 11+j2πfRC
|H (f)| = 11+4π2f21R2C2√
H (0) = 1
Given that |H(f1)|H(0)≥0.95
Or, 1 + 4π2 f12 R2 C2 ≤ 1.108
Simplifying, f1 ≤ 0.3292πRC
∴f1 ≤ 52.2 Hz.

351. The response of the LTI system for d2y(t)dt2+dy(t)dt+5y(t)=dx(t)dt. Given that y(0–) = 2,
dx(t)dt (at t=0) = 0, x(t) = u(t) is __________
a) 2e-t cos t u(t)
b) 0.5 e-t sin t u(t)
c) 2e-t cos t u(t) + 0.5 e-t sin t u(t)
d) 0.5 e-t cos t u(t-1) + 2e-t sin t u(t-1)
Answer: c
Explanation: s2Y(s) – 2s + 2sY(s) – 2 + 5Y(s) = 1
∴ (s2+2s+5) Y(s) = 3+2s
Or, Y(s) = 2s+3s2+2s+5
= 2(s+1)(s+1)2+22+1(s+1)2+22
Hence, y (t) = 2e-t cos t u(t) + 0.5 e-t sin t u(t)

Sampling
352. Find the Nyquist rate and Nyquist interval of sin(2πt).
a) 2 Hz, 12 sec
b) 12 Hz, 12 sec
c) 12 Hz, 2 sec
d) 2 Hz, 2 sec
Answer: a
Explanation: We know that sin⁡ ω0 t ↔ jπ[δ(ω+ω0) – δ(ω-ω0)]
sin⁡ 2πt ↔ jπ[δ(ω+2π)-δ(ω-2π)]
Here ωm = 2π
But ωm = 2πfm
∴ fm = 1 Hz
Nyquist rate, Fs = 2fm = 2 Hz
Nyquist interval, T = 12fm=12sec.

Department of Electrical and Computer Engineering 84

Wollo University Kombolcha Institute of Technology

353. Find the Nyquist rate and Nyquist interval of sinc[t].

a) 1 Hz, 1 sec
b) 2 Hz, 2 sec
c) 12 Hz, 2 sec
d) 2 Hz, 12sec
Answer: a
Explanation: We know that sinc[t] ↔ G2π(ω)
Here ωm = 2π
2πfm = π
∴ 2fm = 1
Nyquist rate, Fs = 2fm = 1 Hz
Nyquist interval, T = 12fm = 1 sec.

354. Find the Nyquist rate and Nyquist interval of Asinc[t].

a) 2 Hz, 2 sec
b) 1 Hz, 1 sec
c) 12 Hz, 1 sec
d) 1 Hz, 12 sec
Answer: b
Explanation: Nyquist rate and Nyquist interval are independent of Amplitude (magnitude
scaling). But time scaling will change the rate.
We know that sinc[t] ↔ G2π(ω)
Here ωm = 2π
2πfm = π
∴ 2fm = 1
Nyquist rate, Fs = 2fm = 1 Hz
Nyquist interval, T = 12fm = 1 sec.
∴Fs = 1 Hz, T = 1 sec.

355. Find the Nyquist rate and Nyquist interval of sinc[200t].

a) 200 Hz, 1200 sec
b) 200 Hz, 200 sec
c) 1200 Hz, 200 sec
d) 100 Hz, 100 sec
Answer: a
Explanation: Here ωm=200π
2πfm=200π
2fm=200 Hz
Nyquist rate, Fs = 2fm = 200 Hz
Nyquist interval, T = 12fm=1200 sec.

356. Which of the following is the process of ‘aliasing’?

a) Peaks overlapping
b) Phase overlapping
c) Amplitude overlapping
d) Spectral overlapping

Department of Electrical and Computer Engineering 85

Wollo University Kombolcha Institute of Technology

Answer: d
Explanation: Aliasing is defined as the phenomenon in which a high frequency component in the
frequency spectrum of the signal takes the identity of a lower frequency component in the
spectrum of the sampled signal.
Aliasing can occur if either of the following condition exists:
• The signal is not band-limited to a finite range.
• The sampling rate is too low.

357. Find the Nyquist rate and Nyquist interval for the signal f(t)=sin500πtπt.
a) 500 Hz, 2 sec
b) 500 Hz, 2 msec
c) 2 Hz, 500 sec
d) 2 Hz, 500 msec
Answer: b
Explanation: Given f(t) = sin500πtπt
Frequency, ωm = 500π
2πfm = 500π
2fm = 500 Hz
Nyquist rate, Fs = 2fm = 500 Hz
Nyquist interval, T = 12fm=1500 = 2 msec.

358. Find the Nyquist rate and Nyquist interval for the signal f(t) = [sin500πtπt]2.
a) 1000 Hz, 1 msec
b) 1 Hz, 1000 sec
c) 1000 Hz, 1 sec
d) 1000 Hz, 1000 sec
Answer: a
Explanation: Given f(t) = [sin500πtπt]2=1−cos1000πt(πt)2
Frequency, ωm = 1000π
2πfm = 1000π
2fm = 1000 Hz
Nyquist rate, Fs = 2fm = 1000 Hz
Nyquist interval, T = 12fm=11000 = 1 msec.

359. Find the Nyquist rate and Nyquist interval for the signal f(t) = 1 + sinc300πt.
a) 300 Hz, 3 msec
b) 300 Hz, 3.3 msec
c) 30 Hz, 3 msec
d) 3 Hz, 3 msec
Answer: b
Explanation: Given f(t) = 1 + sinc300πt
Frequency, ωm = 300π
2πfm = 300π
2fm = 300 Hz

Department of Electrical and Computer Engineering 86

Wollo University Kombolcha Institute of Technology

Nyquist rate, Fs = 2fm = 300 Hz

Nyquist interval, T = 12fm=1300 = 3.3 msec.

360. Find the Nyquist rate and Nyquist interval for the signal f(t) = rect(200t).
a) ∞ Hz, 0 sec
b) 0 Hz, ∞ sec
c) ∞ Hz, ∞ Hz
d) 0 Hz, 0 sec
Answer: a
Explanation: Given f(t) = rect(200t), which is a rectangular pulse signal having pulse width of
1/200 seconds. Since the signal is a finite duration signal, it is not band-limited. The signal
spectrum consists of infinite frequencies.
Hence, Nyquist rate is infinity and Nyquist interval is zero.

361. The sampling frequency of a signal is Fs = 2000 samples per second. Find its Nyquist
interval.
a) 0.5 sec
b) 5 msec
c) 5 sec
d) 0.5 msec
Answer: b
Explanation: Given Fs = 2000 samples per second
Nyquist interval, T = 1Fs=12000 = 0.5 msec.

362. Determine the Nyquist rate of the signal x(t) = 1 + cos⁡ 2000πt + sin⁡ 4000πt.
a) 2000 Hz
b) 4000 Hz
c) 1 Hz
d) 6000 Hz
Answer: b
Explanation: Given x(t) = 1 + cos 2000πt + sin⁡ 4000πt
Highest frequency component in 1 is zero
Highest frequency component in cos⁡2000πt is ωm1 = 2000π
Highest frequency component in sin⁡4000πt is ωm2 = 4000π
So the maximum frequency component in x(t) is ωm = 4000π [highest of 0, 2000π, 4000π]
∴ 2πfm = 4000π
2fm = 4000
Nyquist rate, Fs = 2fm = 4000 Hz.

363. Find the Nyquist rate and Nyquist interval for the signal f(t) = -10 sin ⁡40πt cos ⁡300πt.
a) 40 Hz, 40 sec
b) 340 Hz, 340 sec
c) 300 Hz, 300 sec
d) 340 Hz, 1340 sec
Answer: d
Explanation: sin ⁡40πt has highest frequency ωm1 = 40π
Department of Electrical and Computer Engineering 87
Wollo University Kombolcha Institute of Technology

cos⁡300πt has highest frequency ωm2 = 300π

As we know, multiplication in time domain is equivalent to convolution in frequency domain,
the convoluted spectra will have highest frequency component ωm = ωm1 + ωm2 = 40π + 300π
ωm = 340π
2πfm = 340π
2fm = 340
Nyquist rate, Fs = 2fm = 340 Hz
Nyquist interval, T = 1Fs=1340 sec.

The Laplace Transform

364. The necessary condition for convergence of the Laplace transform is the absolute
integrability of f(t)e-σt.
a) True
b) False
Answer: a
Explanation: The necessary condition for convergence of the Laplace transform is the absolute
integrability of f(t)e-σt.Mathematically, this can be stated as
∫∞−∞|f(t)e−σt|dt<∞
Laplace transform exists only for signals which satisfy the above equation in the given region.

365. Find the Laplace transform of e-at u(t) and its ROC.
a) 1s−a, Re{s}>-a
b) 1s, Re{s}>a
c) 1s×a, Re{s}>a
d) 1s+a, Re{s}>-a
Answer: d
Explanation: Laplace transform, L{x(t)} = X(s) = ∫∞−∞x(t)e−stdt
L{e-at) u(t)} = ∫∞−∞e−atu(t)e−stdt=∫∞0e−ate−stdt=1s+a when (s+a)>0
(σ+a)>0
σ>-a
ROC is Re{s}>-a.

366. Find the Laplace transform of δ(t).

a) 1
b) 0
c) ∞
d) 2
Answer: a
Explanation: Laplace transform, L{x(t)} = X(s) = ∫∞−∞x(t)e−stdt
L{δ(t)} = ∫∞−∞δ(t)e−stdt
[x(t)δ(t) = x(0)δ(t)]

Department of Electrical and Computer Engineering 88

Wollo University Kombolcha Institute of Technology

= ∫∞−∞δ(t)dt
= 1.

367. Find the Laplace transform of u(t) and its ROC.

a) 1s, σ<0
b) 1s, σ>0
c) 1s−1, σ=0
d) 11−s, σ≤0
Answer: b
Explanation: Laplace transform, L{x(t)} = X(s) = ∫∞−∞x(t)e−stdt
L{u(t)} = ∫∞−∞u(t)e−stdt=∫∞0e−stdt=1s when s>0 i.e,σ>0.

368. Find the ROC of x(t) = e-2t u(t) + e-3t u(t).

a) σ>2
b) σ>3
c) σ>-3
d) σ>-2
Answer: d
Explanation: Given x(t) = e-2t u(t) + e-3t u(t)
Laplace transform, L{x(t)} = X(s) = ∫∞−∞x(t)e−stdt
X(s) = 1s+2+1s+3
ROC is {σ > -2}∩{σ > -3}
Hence, the ROC is σ > -2.

369. Find the Laplace transform of cos⁡ωt u(t).

a) ss2+ω2
b) ss2−ω2
c) ωs2+ω2
d) ωs2−ω2
Answer: a
Explanation: Laplace transform, L{x(t)} = X(s) = ∫∞−∞x(t)e−stdt
X(s) = L{cos⁡ωt u(t)} = L[ejωt+e−jωt2u(t)]=12L[ejωtu(t)]+12L[ejωtu(t)]
X(s) = 12(1s−jω)+12(1s+jω)=ss2+ω2.

370. Find the Laplace transform of e-at sin⁡ωt u(t).

a) s+a(s+a)2−ω2
b) ω(s+a)2−ω2
c) s+a(s+a)2+ω2
d) ω(s+a)2+ω2
Answer: d
Explanation: Laplace transform, L{x(t)} = X(s) = ∫∞−∞x(t)e−stdt
L{x(t)} = X(s) = Le−atejωt−e−jωt2ju(t)=12jL[e−(a−jω)tu(t)]–12jL[e−(a+jω)tu(t)]

Department of Electrical and Computer Engineering 89

Wollo University Kombolcha Institute of Technology

X(s) = 12j[1s+(a−jω)–1s+(a+jω)]=12j[2jω(s+a)2+ω2]=ω(s+a)2+ω2
e^-at sin⁡ωt u(t) LT←→ω(s+a)2+ω2;ROC Re(s)>-a.

371. Find the Laplace transform of the signal x(t)=et sin⁡2t for t≤0.
a) 2(s−1)2+22
b) −2(s−1)2+22
c) 2(s+1)2+22
d) −2(s+1)2+22
Answer: b
Explanation: Given x(t) = et sin⁡2t for t≤0
∴ x(t) = et sin⁡2t u(-t)
L{x(t)} = X(s) = ∫∞−∞x(t)e−stdt=∫∞−∞etsin2tu(−t)e−stdt
= ∫0−∞(ej2t–e−j2t2j)=12j∫0−∞[e(1−s+j2)t–e(1−s−j2)t]dt
= 12j(11−s+j2−11−s−j2)
=−2(s−1)2+22.

372. Find the Laplace transform of the signal x(t)=te-2|t|.

a) −1(s−2)2+1(s+2)2
b) 1(s−2)2+1(s+2)2
c) 1(s−2)2–1(s+2)2
d) −1(s−2)2–1(s+2)2
Answer: a
Explanation: Given x(t)=te-2|t|
L{x(t)} = X(s) = ∫∞−∞x(t)e−stdt=∫∞−∞te−2|t|estdt
=∫0−∞te2te−stdt+∫∞0te−2te−stdt=−1(s−2)2+1(s+2)2.

373. Find the Laplace transform of (cos⁡2t)3 u(t).

a) s(s2+28)(s2+36)(s2+4)
b) s(s2+36)(s2+28)(s2+4)
c) s(s2+4)(s2+36)(s2+28)
d) s(s2+36)(s2+4)
Answer: a
Explanation: Given x(t)=(cos⁡2t)3 u(t) = [cos6t+3cos2t]4 u(t)
X(s) = L{x(t)} = L[(cos6t+3cos2t)4u(t)]=14 {L[cos⁡6t u(t)]+3L[cos⁡2t u(t)]}
= 14(ss2+(6)2+3ss2+(2)2)=s(s2+28)(s2+36)(s2+4).

374. Find the Laplace transform of [1 +sin 2t cos ⁡2t]u(t).

a) s2+2s+16s(s2−42)
b) s2+2s+16s(s2+42)
c) s2+2s+16(s2+42)
d) s2+2s+16s

Department of Electrical and Computer Engineering 90

Wollo University Kombolcha Institute of Technology

Answer: b
Explanation: Given x(t)=[1 + sin ⁡2t cos ⁡2t]u(t) = (1 + 12sin4t)u(t)
L{x(t)} = X(s) = L[u(t) + 12 sin⁡4t u(t)] = L[u(t)] + 12 L[sin⁡4t u(t)] =
1s+124(s2+42)=s2+2s+16s(s2+42).

375. Find the Laplace transform of x(t) = u(t+2) + u(t-2).

a) cos2ss
b) cosh2ss
c) sinh2ss
d) sin2ss
Answer: b
Explanation: Given x(t) = u(t+2) + u(t-2)
We know that the Laplace transform u(t) ↔ 1s
Time shifting property states that L{x(t±t0)} = X(s)e±st0
L{u(t-2)}=e−2ss; L{u(t+2)}=e2ss
∴X(s) = L{u(t+2)+u(t-2)} = e−2s+e−2ss=cosh2ss.

376. Find the Laplace transform of the signal x(t) = e-2t cos⁡(200πt)u(t).
a) ss2+(200π)2
b) ss2−(200π)2
c) s−2(s−2)2+(200π)2
d) s+2(s+2)2+(200π)2
Answer: d
Explanation: Given x(t) = e-2t cos⁡(200πt)u(t)
We know that L{cos⁡ωt u(t)} = ss2+ω2
∴L{cos⁡(200πt)u(t)} = ss2+(200π)2
Frequency shifting property states that L{e-at x(t)} = X(s+a)
L{e-2t cos⁡(200πt)u(t)} = L{cos⁡(200πt)u(t)}|s=s+2 = [ss2+(200π)2]s=s+2=s+2(s+2)2+(200π)2.

377. Find the Laplace transform of the signal x(t) = sin⁡(t2)u(t2).

a) 1s2+1
b) ss2+1
c) 2s(2s)2+1
d) 2(2s)2+1
Answer: d
Explanation: We know that sin⁡t u(t) ↔ 1s2+1
Scaling property states that f(at) ↔ 1aF(sa)
sin(t2)u(t2)↔1(12)1[(s1/2)2+1]↔2(2s)2+1.
378. Find the Laplace transform of the signal x(t) = dδ(t)dt.
a) 1
b) s
c) 1s
d) s2

Department of Electrical and Computer Engineering 91

Wollo University Kombolcha Institute of Technology

Answer: b
Explanation: Given x(t) = dδ(t)dt
We know that L{δ(t)} = 1
Time differentiation property, L{dδ(t)dt} = sF(s)
L{dδ(t)dt} = sL{δ(t)} = s × 1 = s.

379. Find the Laplace transform of the signal x(t) = te-αt.

a) 1s2
b) 1(s+α)2
c) 1α
d) 1s+α
Answer: b
Explanation: We know that L{e-αt} = 1s+α
Differentiation in s-domain property states that (-t)n f(t) ↔ dnF(s)dsn
L{te-αt} = –dds[1s+α]=1(s+α)2.

380. Find the Laplace transform for f(t) = 1t [e-2t – e-3t]u(t).

a) ln(s−2s−3)
b) ln(s+2s+3)
c) ln(s−2s+3)
d) ln(s+2s−3)
Answer: b
Explanation: Given f(t) = 1t [e-2t – e-3t]u(t)
We know that L{e-2t) u(t)} = 1s+2; L{-3t u(t)} = 1s+3
Integration in s-domain property states that ∫∞sF(s)ds↔f(t)t
L{1t [e-2t – e-3)]u(t)} = ∫∞s(1s+2–1s+3)ds=[ln(s+2)–ln(s+3)]|∞s=ln(s+2s+3).

381. Find the initial value of f(t) if F(s) = s(s+a)2+ω2.

a) 0
b) -1
c) ∞
d) 1
Answer: d
Explanation: Given F(s) = s(s+a)2+ω2
The initial value is f(0+) = lims→∞ sF(s)
= lims→∞⁡ s s(s+a)2+ω2=lims→∞1(1+a/s)2+(ω/s)2=1.

382. Find the final value of the function F(s) given by (s−1)s(s2−1).
a) 1
b) 0
c) -1
d) ∞
Answer: a
Explanation: Given F(s) = (s−1)s(s2−1)
Department of Electrical and Computer Engineering 92
Wollo University Kombolcha Institute of Technology

The final value is f(∞)=lims→0⁡ sF(s)

= lims→0⁡ s (s−1)s(s2−1)=lims→01s+1=1.

383. Determine the initial value x(0+) for the Laplace transform X(s) = 4s2+3s−5.
a) -1
b) 0
c) 1
d) ∞
Answer: b
Explanation: Given X(s) = 4s2+3s−5
Initial value, x(0+) = lims→∞⁡ sX(s) = lims→∞⁡ s(4s2+3s−5) = limx→0⁡ 4x1+3x−5x2=0 [let s =
1/x].

384. Find x(∞) if X(s) is given by s−2s(s+4).

a) 1
b) -1
c) 12
d) –12
Answer: d
Explanation: Given X(s) = s−2s(s+4)
Final value, x(∞) = lims→0 sX(s) = lims→0 (s−2)s(s+4)=−24=−12.

ROC Properties
385. Given a system function H(s) = 1s+3. Let us consider a signal sin 2t. Then the steady state
response is ___________
a) 18
b) Infinite
c) 0
d) 8
Answer: c
Explanation: H(s) = V(s)J(s)
= 1s+3
V(s) = 1s+3. J(s)
J(s) = L (sin 2t) = 2s2+4
V (s) = 1s2+4.2s+3
VSS = lims→0 sV(s)
= 0.

386. If H(f) = y(t)x(t), then for this to be true x(t) is ___________

a) exp(j2nft)
b) exp(−j2nft)
c) exp(j2nft)

Department of Electrical and Computer Engineering 93

Wollo University Kombolcha Institute of Technology

d) exp(-j2nft)
Answer: c
Explanation: Let us consider x (t) = e2jnft
So, y (t) = ∫∞−∞h(τ)x(t−τ)dτ
= ∫∞−∞h(τ)ej2nπf(t−τ)dτ
= ej2nft H (f)
Or, H (f) = y(t)ej2nft
So, x (t) = ej2nft.

387. The z-transform of –u(-n-1) is ___________

a) 11−z
b) z1−z
c) 11−z−1
d) z1−z−1
Answer: c
Explanation: z [-u (-n-1)] = ∑∞n=−1[u(−n−1)]z−n
= – [z + z2 + z3 + z4…]
= zz−1
= 11−z−1.

388. The value of z(ak u[-k]) is _______________

a) zz−a
b) za−z
c) z2z−a
d) aa−z
Answer: b
Explanation: z (ak u [-k]) = ∑−1k=−∞akz−k
= ∑∞m=1a−mzm
= ∑∞m=1(a−1z)m
= a−1z1−a−1z=za−z.

389. If a system has N different poles, then the system can have ______________
a) N ROC’s
b) (N-1) ROC’s
c) (N+1) ROC’s
d) 2N ROC’s
Answer: c
Explanation: Let us consider 2 poles. For 2 poles, we will have 3 ROC conditions. Hence, if a
system has N poles then the system will have (N+1) ROC’s.

390. Given 2 signals (-3)k u(k) and u (k-1). These two signals are superimposed. This
superimposed signal is _______________
a) zz+3+1z−1

Department of Electrical and Computer Engineering 94

Wollo University Kombolcha Institute of Technology

b) zz+3–1z−1
c) zz−3+1z−1
d) zz+3+1z+1
Answer: a
Explanation: We know that superposition means addition of these 2 signals.
So, superimposed f[k] = (-3)k u(k) + u(k-1)
Hence, z[k] = zz+3+1z−1.

391. X1(z) = 2z + 1 + z-1 and X2(z) = z + 1 + 2z-1 is ________________

a) Even signal
b) Odd signal
c) In time power signal
d) In time energy signal
Answer: d
Explanation: Given X1(z) = 2z + 1 + z-1 and X2(z) = z+1+2z-1.
So, x1(k) = {2, 1, 1}; x2(k) = {1, 1, 2}.

392. The value of z{[k-1] u(k)} is _______________

a) z(z+2)(z−1)2
b) 2z−z2(z−1)2
c) z2(z−1)2
d) z(z−2)(z+1)2
Answer: b
Explanation: z {[k-1] u (k)} = z {k u (k) – u (k)}
= z(z−1)2–zz−1
= z−z(z−1)(z−1)2
= z−z2+z(z−1)2
= 2z−z2(z−1)2.

393. The area under Gaussian pulse ∫∞−∞e−πt2dt is ___________

a) Unity
b) Infinity
c) Pulse
d) Zero
Answer: a
Explanation: Putting πt2 = x, we get,
∫∞−∞e−πt2dt=∫∞−∞e−x2πxπ−−√dx
=2π−−√∫∞−∞x−−√e−xdx
= 1.

394. The system x(k) = 7(13)k u(-k-1)-6(12)k u(k) is ___________

a) Causal
b) Anti-causal

Department of Electrical and Computer Engineering 95

Wollo University Kombolcha Institute of Technology

c) Non-causal
d) Cannot be determined
Answer: c
Explanation: Taking the z-transform, we get,
X (z) = 7(1z−13)–6(1z−12)
∴ the ROC for given condition is as derived above.
∴ the bounded signal as a whole is non-causal.

395. The spectral density of white noise is ____________

a) Exponential
b) Uniform
c) Poisson
d) Gaussian
Answer: b
Explanation: The distribution of White noise is homogeneous over all frequencies. Power
spectrum is the Fourier transform of the autocorrelation function. Therefore, the power spectral
density of white noise is uniform.

396. In the circuit given below, the value of ZL for maximum power to be transferred is
_____________

a) R
b) R + jωL
c) R – jωL
d) jωL
Answer: c
Explanation: The value of load for maximum power transfer is given by the complex conjugate
of ZAB
ZAB = R + jXL
= R + jωL
∴ ZL for maximum power transfer is given by ZL = R – jωL.

Department of Electrical and Computer Engineering 96

Wollo University Kombolcha Institute of Technology

Inverse Laplace Transform

397. Find the inverse Laplace transform for 1(s+1)2.
a) tet u(t)
b) te-t u(t)
c) tu(t)
d) et u(t)
Answer: b
Explanation: Given X(s) = 1(s+1)2
x(t) = L-1 [X(s)] = L−1[1(s+1)2]=e−tL−1[1s2] = e-t tu(t) = te-t u(t).

398. Find the inverse Laplace transform for 1(s+1)2+1.

a) te-t u(t)
b) e-t sin⁡t u(t)
c) e-t cos⁡t u(t)
d) e-t u(t)
Answer: b
Explanation: Given X(s) = 1(s+1)2+1
x(t) = L-1 [X(s)] = L−1[1(s+1)2+1]=e−tL−1[1s2+1] = e-t sin⁡t u(t).

399. Find the inverse Laplace transform for s(s+2)2.

a) te-t u(t)
b) e-t sin⁡t u(t)
c) e-2t (1-2t)u(t)
d) e2t (1-2t)u(t)
Answer: c
Explanation: Given X(s) = s(s+2)2
x(t) = L-1 [X(s)] = L−1[s(s+2)2]=L−1[s+2(s+2)2–2(s+2)2]=L−1[1s+2]–2L−1[1(s+2)2]
= e-2t – 2e-2t L-1 [1s2] = e-2t (1-2t)u(t).

400. Find the inverse Laplace transform for s(s+2)2+1.

a) [2e-2t cos⁡t + e-2t sin⁡t]u(t)
b) [e-2t cos⁡t + 2e-2t sin⁡t]u(t)
c) [2e-2t cos⁡t – e-2t sin⁡t]u(t)
d) [e-2t cos⁡t – 2e-2t sin⁡t]u(t)
Answer: d
Explanation: Given X(s) = s(s+2)2+1
x(t) = L-1 [X(s)] = L−1[s(s+2)2+1]=L−1[s+2(s+2)2+1–2(s+2)2+1]
=L−1[s+2(s+2)2+1]–2L−1[1(s+2)2+1]=e−2tL−1[ss2+1]–2e−2tL−1[1s2+1]
= [e-2t cos⁡t – 2e-2t sin⁡t]u(t).

401. Find the inverse Laplace transform of X(s) = ss2a2+b2.

a) 1a2cos(ab)t

Department of Electrical and Computer Engineering 97

Wollo University Kombolcha Institute of Technology

b) 1a2cos(ba)t
c) 1a2sin(ba)t
d) 1a2sin(ab)t
Answer: b
Explanation: Given X(s) = ss2a2+b2=1a2[ss2+(b/a)2]
We know that L-1 (ss2+ω2) = cos⁡ωt
∴x(t) = L-1 [X(s)] = 1a2L−1[ss2+(b/a)2]=1a2cos(ba)t.

402. Find the inverse Laplace transform of X(s) = s(s2+a2)2.

a) 1a t sin⁡at
b) 12a t sin⁡at
c) 1a t cos⁡at
d) 12a t cos⁡at
Answer: b
Explanation: Given X(s) = s(s2+a2)2
x(t) = L-1 [X(s)] = L−1[s(s2+a2)2]=12a[−dds{as2+a2}]
= 12atL−1[as2+a2]=12atsinat.

403. If F1 (s) = 1s+2 and F2 (s) = 1s+3, find the inverse Laplace transform of F(s) = F1 (s) F2 (s).
a) [e-2t + e-3t]u(t)
b) [e-2t – e-3t]u(t)
c) [e2t + e3t]u(t)
d) [e2t + e-3t]u(t)
Answer: b
Explanation: Given F1 (s) = 1s+2 and F2 (s) = 1s+3.
F(s) = F1 (s) F2 (s) = (1s+2)(1s+3)=1s+2–1s+3
Applying inverse Laplace transform, we get
f(t) = [e-2t – e-3t]u(t).

404 Find the inverse Laplace transform for X(s) = s2s2−8.

a) cosh⁡2t
b) 12 cosh⁡2t
c) sinh⁡2t
d) 12 sinh⁡2t
Answer: b
Explanation: Given X(s) = s2s2−8=12s(s2−22)
We know that cosh⁡ωt = s(s2−w2)
∴x(t)=L-1 [X(s)] = 12L−1(s(s2−22))=12cosh2t.

405. Find the inverse Laplace transform for X(s) = ln(s+as+b).

a) e−at–e−btt
b) e−bt–e−att

Department of Electrical and Computer Engineering 98

Wollo University Kombolcha Institute of Technology

c) e−at+e−btt
d) ebt+e−att
Answer: b
Explanation: Given X(s) = ln(s+as+b)
x(t) = L-1 [X(s)] = L-1 [ln(s+as+b)]
L[x(t)] = ln(s+as+b) = ln⁡(s+a)-ln⁡(s+b)
L[tx(t)] = –dds [ln⁡(s+a)-ln⁡(s+b)] = −1s+a+1s+b=1s+b–1s+a
tx(t) = L−1(1s+b–1s+a) = e-bt – e-at
x(t) = e−bt–e−att.

406. Find the inverse Laplace transform for the function X(s) = 2s−1s2+4s+8.
a) e-2t cos⁡2t u(t) – e-2t sin⁡2t u(t)
b) 2e-2t cos⁡2t u(t) – 52 e-2t sin⁡2t u(t)
c) 2e-2t cos⁡2t u(t) – e-2t sin⁡2t u(t)
d) e-2t cos⁡2t u(t) – 52 e-2t sin⁡2t u(t)
Answer: b
Explanation: Given function X(s) = 2s−1s2+4s+8=2s−1(s+2)2+22=2(s+2)−5(s+2)2+22
= 2(s+2)(s+2)2+22–522(s+2)2+22
Applying inverse Laplace transform, we get
x(t) = 2e-2t cos⁡2t u(t) – 52 e-2t sin⁡2t u(t).

407. Find the inverse Laplace transform for the function X(s) = 1+e−2s3s2+2s.
a) e-(2/3)t u(t) – u(t) + e-(2/3)(t-2) u(t-2)-u(t-2)
b) e-(2/3)t u(t) + e-(2/3)(t-2) u(t-2)
c) e-(2/3)(t-2) u(t-2) – u(t-2)
d) e-(2/3)t u(t) – u(t)
Answer: a
Explanation: Given function X(s) = 1+e−2s3s2+2s
x(t) = L-1 [X(s)] = L−1[1+e−2s3s2+2s]=L−1[13s2+2s]+L−1[e−2s3s2+2s]
L−1[13s2+2s]=L−1{13s[s+(2/3)]}=L−1{−1s+1[s+(2/3)]} = e-(2/3)t u(t) – u(t)
L−1[e−2s3s2+2s]=L−1(13s2+2s)t=t−2 = e-(2/3)(t-2) u(t-2)-u(t-2)
∴x(t) = e-(2/3)t u(t) – u(t) + e-(2/3)(t-2) u(t-2)-u(t-2).

408. Given x(t)=e-t u(t). Find the inverse Laplace transform of e-3s X(2s).
a) 12 e-(t-3)/2 u(t+3)
b) 12 e-(t-3)/2 u(t-3)
c) 12 e(t-3)/2 u(t-3)
d) 12 e(t-3)/2 u(t+3)
Answer: b
Explanation: Given x(t) = e-t u(t)
X(s) = L[x(t)] = L[e-t u(t)] = 1s+1
X(2s) = 12s+1=1/2s+(1/2)

Department of Electrical and Computer Engineering 99

Wollo University Kombolcha Institute of Technology

L-1 [X(2s)] = L−1[1/2s+(1/2)]=112 e-t/2 u(t)

L-1 [e-3s X(2s)] = L-1 [X(2s)]t=t-3 = 12 e-(t-3)/2 u(t-3)
∴L-1 [e-3s X(2s)] = 12 e-(t-3)/2 u(t-3) if x(t) = e-t u(t)

Z-Transform
409. When do DTFT and ZT are equal?
a) When σ = 0
b) When r = 1
c) When σ = 1
d) When r = 0
Answer: b
Explanation: Discrete Time Fourier Transform, X(e-jω) = ∑∞n=−∞x(n)e−jωn
Z-Transform, X(Z) = ∑∞n=−∞x(n)z−n, z = r ejω
When r=1, z = ejω and hence DTFT and ZT are equal.

410. Find the Z-transform of δ(n+3).

a) z
b) z2
c) 1
d) z3
Answer: d
Explanation: Given x(n) = δ(n+3)
We know that δ(n+3) = {10n=−3otherwise
X(Z) = ∑n=−∞∞x(n)z−n=∑n=−∞∞δ(n+3)z−n = z3.

411. Find the Z-transform of an u(n);a>0.

a) zz−a
b) zz+a
c) 11−az
d) 11+az
Answer: a
Explanation: Given x(n) = an u(n)
We know that u(n)={10n≥0n<0
X(Z) = ∑n=−∞∞x(n)z−n=∑n=−∞∞anu(n)z−n
= ∑n=0∞an(1)z−n=∑n=0∞(az−1)n=(1−az−1)−1
= 11−az−1=zz−a.

412. Find the Z-transform of cos⁡ωn u(n).

a) z(z+cosω)z2−2zcosω+1
b) z(z−cosω)z2−2zcosω+1
c) z(z−cosω)z2+2zcosω+1

Department of Electrical and Computer Engineering 100

Wollo University Kombolcha Institute of Technology

d) z(z+cosω)z2+2zcosω+1
Answer: b
Explanation: Given x(n) = cos⁡ωn u(n)
We know that u(n)={10n≥0n<0
Z[cos⁡ωn u(n)] = Z[ejωn+e−jωn2u(n)]=12Z[ejωnu(n)]+12Z[e−jωnu(n)]
=12(zz−ejω+zz−e−jω)=12[z(z−e−jω)+z(z−ejω)(z−ejω)(z−e−jω)]
=12{z[2z−(ejω+e−jω)]z2−z(ejω+e−jω)+1}=z(z−cosω)z2−2zcosω+1.
413. For causal sequences, the ROC is the exterior of a circle of radius r.
a) True
b) False
Answer: a
Explanation: Consider a causal sequence, x(n) = rn u(n)
X(Z) = ∑n=−∞∞x(n)z−n=∑n=−∞∞rnu(n)z−n=∑n=0∞rn(1)z−n=∑n=0∞(rz−1)n
The above summation converges for |rz-1|<1, i.e. for |z|>r
Hence, for the causal sequences, the ROC is the exterior of a circle of radius r.

414. Find the Z-transform of y(n) = x(n+2)u(n).

a) z2 X(Z) – z2 x(0) – zx(1)
b) z2 X(Z) + z2 x(0) – zx(1)
c) z2 X(Z) – z2 x(0) + zx(1)
d) z2 X(Z) + z2 x(0) + zx(1)
Answer: a
Explanation: Given y(n) = x(n+2)u(n)
Y(z) = Z[y(n)] = Z[x(n+2)u(n)] = ∑n=0∞x(n+2)u(n)z−n=∑n=0∞x(n+2)z−n
Let n + 2 = p,i.e.n = p – 2
Y(z) = ∑∞p=2x(p)z−(p−2)=z2∑∞p=2x(p)z−p=z2∑∞p=0x(p)z−p–x(0)–x(1)z−1
=z2 X(Z) – z2 x(0) – zx(1).

415. Find the Z-transform of x(n) = a|n|; |a|<1.

a) zz−a–zz−(1/a)
b) zz−(1/a)–zz−a
c) zz−a+zz−(1/a)
d) 1z−a–1z−(1/a)
Answer: a
Explanation: a^|n| = a^n u(n) + a-n u(-n-1) = an u(n) + (1a)n u(-n-1)
Z[a|n|] = Z[an u(n)] + Z[(1a)n u(-n-1)] = zz−a–zz−(1/a).

416. Find the Z-transform of u(-n).

a) 11−z
b) 11+z
c) z1−z

Department of Electrical and Computer Engineering 101

Wollo University Kombolcha Institute of Technology

d) z1+z
Answer: a
Explanation: Given x(n) = u(-n)
Z[x(n)] = X(Z) = ∑∞n=−∞x(n)z−n=∑∞n=−∞u(−n)z−n=∑0n=−∞(1)z−n
=∑∞n=0zn=11−z.

417. For a right hand sequence, the ROC is entire z-plane.

a) True
b) False
Answer: b
Explanation: If x(n) is finite duration causal sequence (right-sided sequence),the X(z) converges
for all values of z except at z = 0. Hence, the ROC is entire z-plane except at z = 0.

Inverse Z-Transform
418. Given the z-transform pair 3nn2 u[n] ↔ X (z). The time signal corresponding to X(2z) is
___________
a) n23n u[2n]
b) (−32)nn2u[n]
c) (32)nn2u[n]
d) 6nn2u[n]
Answer: c
Explanation: Y (z) = X (2z) ↔ y[n] = 12n x[n]
Or, y[n] = 12n n2 3n u[n]
Or, y[n] = (32)nn2u[n].

419. Given the z-transform pair 3nn2 u[n] ↔ X (z). The time signal corresponding to X(z-1) is
___________
a) n23-nu[n]
b) n23-nu[-n]
c) 1n231nu[n]
d) 1n231nu[−n]
Answer: b
Explanation: Y (z) = X (1z) ↔ y[n] = X [-n]
Or, y[n] = n23-nu[-n].

420. Given the z-transform pair 3nn2 u[n] ↔ X (z). The time signal corresponding to dX(z)dz is
___________
a) (n-1)33n-1u[n-1]
b) n33nu[n-1]
c) (1-n)33n-1u[n-1]
d) (n-1)33n-1u[n]
Answer: c

Department of Electrical and Computer Engineering 102

Wollo University Kombolcha Institute of Technology

Explanation: Y (z) = dX(z)dz

= −z−1[−zdX(z)dz]
Now, Y (z) ↔ y (n) = – (n-1) X [-n-1] Or, y (n) = – (n-1)33n-1u[n-1].

421. Given the z-transform pair 3nn2 u[n] ↔ X (z). The time signal corresponding to z2−z−22 X(z)
is ___________
a) 12(x[n+2]-x[n-2])
b) (x[n+2]-x[n-2])
c) 12(x[n-2]-x[n+2])
d) (x[n-2]-x[n+2])
Answer: a
Explanation: Y (z) = z2−z−22 X (z) ↔ y[n] = 12(x[n+2]-x[n-2]).

422. Given the z-transform pair 3nn2 u[n] ↔ X (z). The time signal corresponding to {X(z)}2 is
___________
a) {x[n]}2
b) x[n]*x[n]
c) x[n]*x[-n]
d) x[-n]*x[-n]
Answer: b
Explanation: Y (z) = X (z) H (z)
Y (z) = X (z) X (z) ↔ y[n] Or, y [n] = x[n]*x[n]

423. The system described by the difference equation y(n) – 2y(n-1) + y(n-2) = X(n) – X(n-1)
has y(n) = 0 and n<0. If x (n) = δ(n), then y (z) will be?
a) 2
b) 1
c) 0
d) -1
Answer: c
Explanation: Given equation = y (n) – 2y (n-1) + y (n-2) = X (n) – X (n-1) has y (n) = 0
For n = 0, y (0)2y (-1) + y (-2) = x (0) – x (-1)
∴ y(0) = x(0) – x(-1)
Or, y (n) = 0 for n<0
For n=1, y (1) = -2y (0) + y (-1) = x (1) – x (0)
Or, y (1) = x (1) – x (0) + 2x (0) – 2x (-1)
Or, y (1) = x (1) + x (0) – 2x (-1)
For n=2, y (2) = x (2) – x (1) + 2y (1) – y (0)
Or, y(2) = x(2) – x(1) + 2x(1) + 2x(0) – 4x(-1) – x(0) + x(-1)
∴y (2) = d (2) + d (1) + d (0) – 3d (-1).

424. The value of Z−1{z2(z−a)(z−b)} is ____________

a) an+1–bn+1a+b
b) an+1–bn+1a−b
c) an+1+bn+1a−b
d) an+1+bn+1a+b
Department of Electrical and Computer Engineering 103
Wollo University Kombolcha Institute of Technology

Answer: b
Explanation: We know that, Z−1zz−a=an andZ−1zz−b=bn
∴Z−1{z2(z−a)(z−b)}=Z−1zz−a.zz−b=an∗bn
= ∑nm=0am.bn−m
= bn∑nm=0amb
= bn.an+1b−1ab−1
= an+1–bn+1a−b.

425. Given the z-transform pair

X[n]↔32z2−16, |z|<4
The z-transform of the signal y[n] = nx[n] is _________
a) 32z2(z2−16)2
b) −32z2(z2−16)2
c) 32(z2−16)2
d) −32z(z2−16)2
Answer: a
Explanation: y[n] = n x[n] n x[n] ↔ Y (z) = −zdX(z)dz
Now, −zdX(z)dz=32z2(z2−16)2.

426. Given the z-transform pair

X[n]↔32z2−16, |z|<4
The z-transform of the signal y[n] = x[n+1] + x[n-1] is _________
a) (z+1)2(z+1)2−16+(z−1)2(z−1)2−16
b) z2(1+z)z2−16
c) z2(z−1)z2−16
d) (z+2)2(z+2)2−16
Answer: b
Explanation: x (n-n0) ↔ z-n0 X (z)
Now, y[n] = x [n+1] + x [n-1] ↔ Y (z)
Y (z) = (z+z-1) X (z)
∴ Y (z) = z2(1+z)z2−16.

427. The value of inverse Z-transform of log(zz+1) is _______________

a) (-1)n/n for n = 0; 0 otherwise
b) (-1)n/n
c) 0, for n = 0; (-1)n/n, otherwise
d) 0
Answer: c
Explanation: Putting z = 1t, U (z) = log (1y1y+1)
= – log (1+y) = -y + 12 y2 – 13 y3 + …..
= -z-1 + 12 z-2 – 13 z-3 + …..
Thus, un = 0, for n = 0; (-1)n/n otherwise

Department of Electrical and Computer Engineering 104

Wollo University Kombolcha Institute of Technology

Convolution
428. The resulting signal when a continuous time periodic signal x(t) having period T, is
convolved with itself is ___________
a) Non-Periodic
b) Periodic having period 2T
c) Periodic having period T
d) Periodic having period T/2
Answer: c
Explanation: The solution lies with the definition of convolution. Given a periodic signal x (t)
having period T. When convolution of a periodic signal with period T occurs with itself, it will
give the same period T.

429. Convolution of step signal 49 times that is 49 convolution operations. The Laplace
transform is ______________
a) 1s49
b) 1s50
c) 1
d) s49
Answer: a
Explanation: n times = u (t) * u (t) * …… * u (t)
Laplace transform of the above function = 1sn, where n is number of convolutions.
∴ Laplace transform for 49 convolutions = 1s49.

430. The auto correlation of x(t) = e-atu(t) is ________________

a) e−ata2
b) e−at2a
c) e−aλa2
d) e−aλ2a
Answer: d
Explanation: R (λ) = ∫∞−∞x(t)x(t±λ)dt
= ∫∞−∞e−atu(t)e−a(t−λ)u(t−λ)dt
= ∫∞λe−2ateaλdt
= eaλ−2a[0-e-2aλ]
= e−aλ2a.

431. For any given signal, average power in its 6 harmonic components as 10 mw each and
fundamental component also has 10 mV power. Then, average power in the periodic signal is
_______________
a) 70
b) 60
c) 10
d) 5
Answer: b

Department of Electrical and Computer Engineering 105

Wollo University Kombolcha Institute of Technology

Explanation: We know that according to Parseval’s relation, the average power is equal to the
sum of the average powers in all of its harmonic components.
∴ Pavg = 10 × 6 = 60.

432. One of the types of signal is an Impulse train. The type of discontinuity in an impulse train
is ______________
a) Infinite
b) Zero
c) One
d) Finite
Answer: a
Explanation: From any Impulse train waveform, we can infer that it is a kind of signal having
infinite discontinuity.

433. Given a signal f (t) = 3t2+2t+1, which is multiplied by 2 unit delayed version of impulse and
integrated over period -∞ to ∞. The resultant is ______________
a) 1
b) 6
c) 17
d) 16
Answer: c
Explanation: ∫∞−∞f(t)δ(t−t0)=f(t0)
Here, t0 = 2
So, ∫∞−∞f(t)δ(t−2) = f (2)
Hence, f (2) = 3(2)2 + 2(2) + 1
= 12 + 4 + 1 = 17.

434. A PT is a device which is ___________

a) Electrostatically coupled
b) Electrically coupled
c) Electromagnetically coupled
d) Conductively coupled
Answer: c
Explanation: A PT cannot be electrostatically coupled since CRO are electrostatically coupled.
Also, they cannot be conductively coupled. But since they are kind of electrically coupled hence
electromagnetically coupled is the only correct option.

435. The CT supplies current to the current coil of a power factor meter, energy meter and, an
ammeter. These are connected as?
a) All coils in parallel
b) All coils in series
c) Series-parallel connection with two in each arm
d) Series-parallel connection with one in each arm
Answer: b
Explanation: Since the CT supplies the current to the current coil, therefore the coils are
connected in series so that the current remains the same. If they were connected in parallel then
Department of Electrical and Computer Engineering 106
Wollo University Kombolcha Institute of Technology

the voltages would have been same but the currents would not be the same and thus efficiency
would decrease.

436. If a signal f(t) has energy E, the energy of the signal f(100t) is equal to ____________
a) E
b) 100E
c) E/100
d) 400E
Answer: c
Explanation: We know that, E = ∫∞−∞f(t)2dt
Let, Es = ∫∞−∞f(t)2dt
Let 100t = p
Or, dt = dp/100
= ∫∞−∞f(t)2dp/100
So, Es = E/100.

437. Two sequences x1 (n) and x2 (n) are related by x2 (n) = x1 (- n). In the z-domain, their region
of convergences are _______________
a) The same
b) Reciprocal of each other
c) Negative of each other
d) Complementary
Answer: b
Explanation: x1(n) has z-transform X1(z)
The ROC = Rx (say)
Again, x2(n) = x1(-n) has z-transform X1(1/z)
The ROC = 1/Rx
Hence they are reciprocals.

438. If the Laplace transform of f (t) = ws2+w2. The value of limt→∞ f(t) is ____________
a) Cannot be determined
b) Zero
c) Unity
d) Infinity
Answer: b
Explanation: We know that,
By final value theorem, limt→∞⁡ f(t) = lims→0 s F (s)
= lims→0 s.ws2+w2
= 0.

Department of Electrical and Computer Engineering 107

Wollo University Kombolcha Institute of Technology

439. The auto-correlation function of a rectangular pulse of duration T is _____________

a) A rectangular pulse of duration T
b) A rectangular pulse of duration 2T
c) A triangular pulse of duration T
d) A triangular pulse of duration 2T
Answer: d
Explanation: Rxx1 = 1T∫T/2−T/2x(t)x(t+T)dt
Which when plotted is a triangular pulse of duration 2T.

440. The power in the signal (t) = 8cos (20πt – π2) + 4sin (15πt) is equal to ______________
a) 40
b) 42
c) 41
d) 82
Answer: a
Explanation: Power of Signal = limT→ ∞ 1T∫T/2−T/2|f(t)|2dt
Signal power P is mean of the signal amplitude squared value of f (t).
Rms value of signal = P−−√
Now, (t) = 8cos (20πt – π2) + 4sin (15πt)
= 8 sin (20πt) + 4 sin (15πt)
= 822+422
= 32 + 8 = 40.

441. The Fourier transform (FT) of a function x (t) is X (f). The FT of dx(t)dt will be
___________
a) dX(f)df
b) 2πjf X(f)
c) X(f) jf
d) X(f)jf
Answer: b
Explanation: x(t)→12π∫∞−∞X(f)ej2πtdt
Now, differentiating both sides,
We get, dx(t)dt=j2π12π∫∞−∞X(f)ej2πtdt
= j2πf X(f).

442. Given the signal

X (t) = cos t, if t<0
X (t) = Sin t, if t≥0
The correct statement among the following is?
a) Periodic with fundamental period 2π
a) Periodic but with no fundamental period
a) Non-periodic and discontinuous
a) Non-periodic but continuous
Answer: c

Department of Electrical and Computer Engineering 108

Wollo University Kombolcha Institute of Technology

Explanation: From the graphs of cos and sin, we can infer that at t=0, the function becomes
discontinuous.
Since, cos 0 = 1, but sin 0 = 0
As 1 ≠ 0, so, the function X (t) is discontinuous and therefore Non-periodic.

Department of Electrical and Computer Engineering 109