SIGNAL AND SYSTEM

WITH GNU OCTAVE

SIMULATION

FAUZIAH BINTI ALIMAN

SYAMSUL BAHRI MOHAMAD
©ePembelajaran Politeknik Merlimau

Writer
Fauziah Binti Aliman
Syamsul Bahri Mohamad

Published in 2022

All rights reserved. No part of this publication may be reproduced, distributed, or transmitted in any
form or by any means, including photocopying, recording, or other electronic or mechanic methods,
without the prior written permission of the writer, except in the case of brief quotations embodied
in reviews and certain other non-commercial uses.

Perpustakaan Negara Malaysia Cataloguing-in-Publication Data

Fauziah Aliman
SIGNAL AND SYSTEM: WITH GNU OCTAVE SIMULATION / FAUZIAH BINTI
ALIMAN, SYAMSUL BAHRI MOHAMAD.
Mode of Access: Internet
eISBN 978-967-2762-14-0
1. Signal processing.
2. Signal theory (Telecommunication).
3. System analysis.
4. Government publications--Malaysia.
5. Electronic books.
I. Syamsul Bahri Mohamad.
II. Title.
621.3822

Published by:
Politeknik Merlimau, Melaka
KB1031 Pej Pos Merlimau,
77300 Merlimau Melaka

3
 EDITORIAL BOARD

MANAGING EDITOR
Ts. Dr Maria binti Mohamad
Sr. Firhan bin Salian
Ts. Rodzah binti Hj Yahya
Aylin binti Kamaruddin
Hafidah binti Mahat
Noraini binti Ya'cob

EDITOR
Shaidzwan Bin A.Rahim
Ketua Program
Diploma Kejuruteraan Elektronik (Komunikasi)

DESIGNER
Fauziah Binti Aliman
Syamsul Bahri Mohamad

PROOFREADING & LANGUAGE

Rozainah binti Abd Latif
Yeo Li Min
Putra Shazly bin Rosman
Gan Ek Hern
Rosheela Binti Muhammad Thangaveloo

4
 Acknowledgement

In the name of Allah, the Most Gracious and the Most Merciful.

My humblest gratitude to the Holy Prophet Muhammad (Peace be upon him) whose
way of life has been continuous guidance for us. Thank you to everyone who has been
supporting us in preparing this e-book. Never stop learning !

52
Preface

This book includes a thorough study of continuous and discrete time

signals and systems, as well as several GNU OCTAVE examples. It is
intended for junior and seniorelectronic engineering students, as well as
working professionals who wish to study independently. A fundamental
course in differential and integral calculus, as well as basic electric circuit
theory, are required.

This book is suitable for one semester course. The author has taught the
subject material at Polytechnic Merlimau in Melaka, Malaysia, for many
years and has covered all of the contents in 14 weeks, with two lecture
hours weekly. GNU Octave is open- source software that can be freely
distributed. Students may redistribute and/or modifyit under the
provisions of the Free Software Foundation's GNU General Public License
(GPL).

In Chapter 1, the fundamental signals are covered, supported with

various examples. This chapter's goal is to teach the reader how to
express any continuous-time or discrete- time waveform in terms of the
unit step function, unit impulse function, and subsequent operation. The
convolution integral for continuous-time signals and the convolution sum
for discrete-time signals are discussed in Chapters 2. The Laplace
Transform and continuous-time signals are introduced in Chapter 3. The
discrete-timesignals and the Z transform are discussed in Chapter 4. The
Fourier Transform and FastFourier Transform (FFT) are covered in
Chapters 5 and 6, with the most basic explanations available.

2
 Table of Contents
01 Signals and Systems Page 1

02 Page 15
Linear Time-invariant Systems

03 Laplace Transform and Continuous-time LTI Systems Page 27

04 Z Transform and Discrete-time LTI Systems Page 35

05 Fourier Analysis of Continuous-time Page 42

06 Fast Fourier Transform Page 48

 Chapter 1 : Signals and Systems

Signals and
Systems
Signals and systems theory and idea are Time is considered as a discrete variable
required in practically all electrical in discrete time, which regards values of
engineering specialties, as well as many variables as occurring at distinct, separate
other technical and scientific disciplines. "points in time," or equivalently as
The mathematical description and remaining unchanging across each non-
representation of signals and systems, as zero region of time.
well as their classifications, are covered in
this chapter along with a few key As time passes from one time period to
fundamental signals that are critical to the the next, a non-time variable jumps from
studies. one value to the next. This perception of
time is analogous to a digital clock that
Discrete time and continuous time are displays a fixed reading, for example, of
two different frameworks for modelling 10:37 for a period of time before jumping
variables that change over time in to a new fixed reading of 10:38, and so
mathematical dynamics. on. In most cases, measurements are
taken at sequential integer values of the
variable time.

Source: pixabay

SIG N AL A N D SYST EM W IT H G N U O C T A VE
1
SIMU L A T IO N
 Chapter 1 : Signals and Systems

Classification of Signal
A signal is a function that represents a In short, a signal x(t) is a continuous-time
physical quantity or variable and usually signal if t is a continuous variable.
incorporates information about the
phenomenon's behavior or nature. In an
resistor-capacitor (RC) circuit, the signal
could represent the voltage across the
capacitor or the current flowing through
the resistor, for example. A signal is
mathematically expressed as a function of
the independent variable t. t usually
stands for time. Thus, a signal is denoted
Fig. 1.1: Graphical representation of continuous-
by x(t).
time signal
Continuous-time Signal

Continuous time considers variables to Basic Continuous-time Signal

have a specific value for an infinitesimally
short period of time. There are an endless A. Unit Step Function
number of points in time between any
two points in time. The unit step function u(t) , also known as
the Heaviside unit function , is defined as
The variable "time" can span the full real
number line or a subset of it, such as the
non-negative reals, depending on the
context. As a result, time is regarded as a
continuous variable.

A continuous signal, also known as a

continuous-time signal, is a variable
quantity (a signal) with a continuous
domain, which is generally time (e.g., a
connected interval of the reals). That is,
the domain of the function is an
uncountable set. The functions Fig. 1.2: Unit Step Function
themselves do not have to be continuous.

2
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

Source: pixabay

Similarly, the shifted unit step function

u(t – t0) is defined as

Fig. 1.4: Unit impulse function

Similarly, the delayed delta function δ(t – t0)

is shown in Fig.1.5
Fig. 1.3: Shifted Unit Step Function

B. Unit Impulse Function

The unit impulse function (t), also known

as the Dirac delta function, plays a central
role in system analysis.

Fig. 1.5: Shifted Unit impulse function

SIG N AL A N D SYST EM W IT H G N U O C T A VE 3
SIMU L A T IO N
 Chapter 1 : Signals and Systems

C. Sinusoidal Signals A sample rate is assigned to a discrete-

time signal acquired by sampling a
A continuous-time sinusoidal signal can sequence at evenly spaced periods.
be expressed as
Discrete-time signals can come from a
x(t) = A cos (ω0t + θ) variety of places, but they usually fall into
one of two categories:
where • By obtaining analogue signal values at
A is the amplitude (real) a fixed or variable pace. This is
ω0 is the radian frequency in referred to as sampling.
radians per second • By looking at a process that is
θ is the phase angle in radians. intrinsically discrete in time, such as
the weekly peak value of a specific
economic indicator.

In short, discrete-time signal x[n] may be

obtained by sampling a continuous-time
signal x(t). Since a discrete-time signal is
defined at discrete times, a discrete-time
signal is often identified as a sequence of
number denoted by x[n], where n is
integer.
Fig. 1.6 : Continuous-time sinusoidal signal.

Discrete-time Signal

A discrete signal, also known as a

discrete-time signal, is a time series made
up of a succession of values. A discrete-
time signal, unlike a continuous-time
signal, does not have a continuous
argument; nonetheless, it can be created
by sampling from a continuous-time
signal.

Fig. 1.7: Graphical representation of discrete-time

signal

SIG N AL A N D SYST EM W IT H G N U O C T A VE 4
SIMU L A T IO N
 Chapter 1 : Signals and Systems

Source: pixabay

Basic Discrete-time Signal

A. Unit Step Sequence

The unit step sequence u[n] is defined as

Fig. 1.9: Shifted unit step sequence

In short, a signal x(t) is a continuous-time
signal if t is a continuous variable.

B. Unit Impulse Sequence

The unit impulse (or unit sample)

sequence δ[n] is defined as

Fig. 1.8: Unit step sequence

Similarly, the shifted unit step sequence

u[n – k] is defined as

5
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

Fig. 1.10 : Unit impulse sequence

Similarly, the shifted unit impulse (or

sample) sequence δ[n – k] is defined as

Fig. 1.12: Example of sinusoidal sequence

Fig. 1.11 : Shifted unit impulse sequence

C. Sinusoidal Sequence

A sinusoidal sequence can be expressed

as
x[n] = A cos (Ωon + θ)

6
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

Source: pixabay

Simulation
Students can learn how to utilize the Basic Continuous-time Signal
octave-online software in order to create
numerous waveforms that are commonly A. Unit Step Function
used for continuous time signal
processing and discrete time signal The unit steps are already incorporated
processing. GNU Octave, an open-source into GNU Octave. The names of the
alternative to MATLAB, has a web mathematicians who utilized them in
interface called Octave Online. their work are used to identify them.
Heaviside(t) is the name of the unit step
function. The example below shows how
they can be used.

t=-20:0.001:20
y=heaviside(t) % Create unit step function
plot(t,y) % Plot unit step function
axis([-10 10 -2 2]) % Define axis margin

7
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

GNU Octave displays the unit step function waveform as shown in Fig. 1.13 where it
satisfies the definition of unit step function as

Fig. 1.13: Unit step function waveform

B. Unit Impulse Function

Similarly , the unit impulse function waveform is illustrated using the below example :

t=-10:10 % Set column

figure(1) % Create a new figure window for plotting
x=[t==0] % Set impulse at t=0
plot(t,x) % Create simple x-y plots

8
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

GNU Octave display unit impulse function waveform as shown in Fig. 1.14 where it
satisfies the definition of unit impulse function as

Fig. 1.14: Unit step function waveform

C. Sinusoidal Signals

The continuous-time sinusoidal waveform signals are illustrated using the following array
commands

t=0:0.01:2; % set axis

f=2; % frequency
a=1; % amplitude
y=a*sin(2*pi*f*t); % define y as sinusoidal function
subplot(2,1,1); % define subplot
plot(t,y) % plot waveform
xlabel('t’); % x axis title
ylabel('y(t)’); % y axis title
title('Sinusoidal C-T Signal’) % plot title

9
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

GNU Octave displays continuous-time sinusoidal signal waveform as shown in Fig. 1.15.

Fig. 1.15: Sinusoidal signal

Basic Discrete-time Signal

A. Unit Step Sequence

The unit step sequence waveform signals are illustrated using the following array
commands

clc % Clear the terminal screen and move the cursor to

the upper left corner.
clear all % Clear all local and global user-defined variables
and all functions from the symbol table.
n = -5 : 5; % set n values
y = heaviside(n); % unit step
stem(n, y) % plot unit step
axis([-5 5 -1 2]) % set axis
xlabel('n’); % set x axis title
ylabel('u[n]’); % set y axis title

title({' Discrete Unit Step Function’}) % set waveform title

10
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

GNU Octave displays the unit step sequence waveform as shown in Fig. 1.16 where it
satisfies the definition of unit step sequence as

Fig. 1.16: Unit step sequence waveform

B. Unit Impulse Sequence

Similarly , the unit impulse sequence waveform is illustrated using below example :

clc % clear the terminal screen and move the cursor to

the upper left corner.
n=-4:4; % set n values
delta_n=[0,0,0,0,1,0,0,0,0] % set unit impulse sequence
stem(n,delta_n) % plot

11
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

GNU Octave displays the unit impulse sequence waveform as shown in Fig. 1.17 where it
satisfies the definition of unit impulse sequence as

Fig. 1.17 : Unit impulse sequence waveform

C. Sinusoidal Sequence

The discrete-time sinusoidal waveform is illustrated using the following array commands

clc % clear the terminal screen

t=0:0.05:2; % set axis
f=2; % frequency
a=1; % amplitude
y=a*sin(2*pi*f*t); % define y as sinusoidal sequence
subplot(2,1,1); % define subplot
stem(t,y) % plot waveform
xlabel('n’); % x axis title
ylabel('y[n]’); % y axis title
title({' Discrete-Time Sinusoidal waveform signal ’}) % plot title

12
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

GNU Octave displays the discrete-time sinusoidal waveform as shown in Fig. 1.18.

Fig. 1.18 : Discrete-time sinusoidal waveform

13
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 1 : Signals and Systems

Source: pixabay

Exercises
1. Write the discrete-time cosine array commands . Use subplot (2,1,2). Label for each axis
and name the title of the graph.

2. Write the array commands for Impulse Shifted Function when δ (n- 2). Label for each
axis and name the title of the graph.

3. Write the waveform command for Step Shifted Function when u (n + 2). Label for each
axis and name the title of the graph.

14
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Linear Time-invariant
(LTI) Systems
Linearity and time-invariance are two of The significance of the convolution
the most significant characteristics of operation in LTI systems arises from the
systems. The fundamental input-output fact that understanding an LTI system's
relationship for systems with these reaction to the unit impulse input helps
features is developed in this chapter. The us to identify its output to any input
input-output relationship for LTI systems signals.
will be shown as to be defined by a
convolution operation.

Source: pixabay

15
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Continuous-Time LTI System and

Convolution Integral
Convolution of two continuous-time Example:
signals x(t) and h(t) is denoted by
Evaluate y(t) = x(t) * h(t), where x(t) and
∞ h(t) are shown in Fig. 2.1, by graphical
y(t) = x(t)*h(t) = ‫׬‬−∞ 𝑥 𝜏 ℎ 𝑡 − 𝜏 𝑑𝜏 method.

It is commonly called the convolution

integral. The convolution integral
operation involves the following four
steps:
a. The impulse response h(τ) is time- Fig. 2.1
reversed (that is, reflected about the
origin) to obtain h (–τ) and then Solution :
shifted by t to form h(t –τ) = h[–(τ –
t)], which is a function of τ with Step 1
parameter t.
b. The signal x(τ) and h(t – τ) are
multiplied together for all values of τ
with t fixed at some values.
c. The product x(τ)h(t – τ) is integrated
over all τ to produce a single output
value y(t).
d. Steps 1 to 3 are repeated as t varies
over – ∞ to ∞ to produce the entire
output y(t).

16
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Step 2 For 5 < t

For t < 0

y(t) = 0 y(t) = 0

For 0 < t < 2 Step 3

Thus, we obtain

0,
𝑡<0
𝑡, 0<𝑡≤2
𝑓 𝑥 = 2, 2<𝑡 ≤3
t 5 − 𝑡, 3<𝑡≤5
y(t) = ‫׬‬0 dτ = t
0, 5<𝑡
For 2 < t < 3

t
y(t) = ‫׬‬t−2 dτ = 2

For 3 < t < 5

3
y(t) = ‫׬‬t−2 dτ = 5 - t

17
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Discrete-Time LTI System and

Convolution Sum
Convolution sum is the convolution of
two sequences x[n] and h[n] which is
denoted by

y[n] = x[n]*h[n]= σ∞
𝑘=−∞ ℎ 𝑘 𝑥[𝑛 − 𝑘]

The convolution sum operation involves Fig. 2.2

the following four steps: Solution :
a. The impulse response h[k] is time-
reversed (that is, reflected about the Step 1
origin) to obtain h[–k] and then Sketch the sequences h[k], x[k] and h[n –
shifted by n to form h[n – k] = h[ – (k k], x[k] h[n – k] for different values of n
– n)], which is a function of k with
parameter n.
b. Two sequences x[k] and h[n – k] are
multiplied together for all values of k
with n fixed at some values.
c. The product x[k]h[n – k] is summed
over all k to produce a single output
sample y[n].
d. Steps 1 to 3 are repeated as n varies For n < 0
over – ∞ to ∞ to produce the entire
output y[n].

Example:

Evaluate y[n] = x[n] * h[n], where x[n] and

h[n] are shown in Fig.2.2 by a graphical y[<0] = 0
method.

18
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Discrete-Time LTI System and

Convolution Sum
For n = 0 For n = 4

y[0] = 1 y[4] = 1 + 1 = 2

For n = 1 For n = 5

y[1] = 1 + 1 = 2 y[5] = 1

For n = 2 For n > 5

y[2] = 1 + 1 + 1 = 3 y[>5] = 0

For n = 3

y[3] = 1 + 1 + 1 = 3

19
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Discrete-Time LTI System and

Convolution Sum
Step 2 or

Thus, summing x[k] h[n – k] for 0 ≤ n ≤ 5, y[n] = {1,2,3,3,2,1}

we obtain

y[0] = 1 y[1] = 2 y[2] = 3

y[3] = 3 y[4] = 2 y[5] = 1

20
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Simulation
Students can learn how to utilize octave-online software to write and execute convolution
integral signals problem as well as convolution sum signals problem.

Convolution Integral using Heaviside Commands

We are utilizing the heaviside(t) command in GNU Octave to perform convolution integral
simulation. To simulate y(t) = x(t) * h(t), two unit step function are used as shown in Fig. 2.3.
The example of command array given shows how the convolution integral is done .

Fig. 2.3 : Input x(t) and impulse response h(t)

% set x(t)
t=-5:0.01:5;
x=heaviside(t)-heaviside(t-1);
subplot(3,1,1);
plot (t,x);
axis([-2 5 -1 1.5]);
xlabel('t'); ylabel('x(t)=u(t)-u(t-1)');
title({'Convolution integral y(t)=x(t)*h(t)'})
% set h(t)
h=heaviside(t)- heaviside(t-4);
subplot(3,1,2);
plot (t,h);
axis([-2 5 -1 1.5]);
xlabel('t'); ylabel('h(t)=u(t)-u(t-4)');
% perform convolution of x(t)*h(t)
m=conv(x,h); subplot(3,1,3); plot (m);
axis([0 2000 -100 150]); xlabel('t'); ylabel('y(t)');

21
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

GNU Octave displays the continuous-time input x(t), impulse response h(t) and
convolution output y(t) as shown in Fig. 2.4.

Fig. 2.4: Continuous-time input x(t), impulse

response h(t) and convolution output y(t)

22
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Convolution Sum

The convolution operation in GNU Octave is performed by using conv command. Similar
to convolution integral, convolution sum is perform on the input x[n] and impulse
response h[n]. The output is denoted by y[n]. The simulation on convolution sum of x[n]
and h[n] can be achieved by using the following command array example. Note that the
bold sequence is referring to n=0.

1 2 4 5
x[n] = ൜0, , , 1, , , 2ቅ
3 3 3 3

h[n] = {1, 1, 1 ,1 , 1}

clear all
%input signal
n=0:6;
xn=0:1/3:2;
figure, stem(n,xn)
grid
xlabel('n’)
ylabel('x[n]’)
title('Input signal')

%impulse response
h1=-2;xh1=1;
h2=-1;xh2=1;
h3=0;xh3=1;
h4=1;xh4=1;
h5=2;xh5=1;
h=[h1,h2,h3,h4,h5];
hn=[xh1,xh2,xh3,xh4,xh5];
figure, stem(h,hn)
grid
xlabel('n’)
ylabel('h[n]’)
title('Impulse Response')

23
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

%output signal
yn=-2:8;
y=conv(hn,xn);
figure, stem(yn,y)
grid
xlabel('n')
ylabel('y[n]')
title('Output 1 of Convolution Sum')

GNU Octave displays the discrete-time input x[n], impulse response h[n] and convolution
output y[n] as shown in Fig. 2.5 to Fig. 2.7.

Fig. 2.5: Discrete-time input x[n]

24
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Fig. 2.6: Discrete-time impulse response h[n]

Fig. 2.7: Convolution Sum output y[n]

25
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 2 : Linear Time-invariant (LTI) Systems

Source: pixabay

Exercises
1. Display the output graph of y(t) = x(t)*h(t) using convolution integral where
x(t) = 3u(t) – 3u(t - 1) and h(t) = 2u(t) – 3u(t - 4) . Write and execute the command.
Choose a suitable axis for all graphs x(t), h(t) and y(t).

2. Write and compute the output of LTI systems for the following signal below for
y[n]=x[n]*h[n] using octave online. Use the function conv or [y=conv(h,x)] to compute
the problem in convolution sum.

x[n] = {1,-1,2,-1,1}
h[n] = {1,1,1,1}

26
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 3 : Laplace Transform and Continuous -time LTI Systems

Laplace Transform and

Continuous-time LTI
Systems
Laplace Transform is to convert time- These rational transforms appear as
domain signals into complex s-domain system functions for LTI systems that
representations in order to analyze and meet linear constant coefficient
process the continuous-time signals and differential equations. The roots of
system. polynomials N(s) and D(s), also known as
zeros and poles, entirely determine
The Laplace transform of a signal x(t) is rational transforms up to a scaling factor.
represented by As these roots are so crucial in studying
+∞ LTI systems, it is indeed easier to visualize
𝑋 𝑠 = න 𝑥(𝑡)𝑒 −𝑠𝑡 them on a pole-zero diagram.
−∞
𝑁(𝑠)
A valuable tool for analyzing linear time- 𝑋 𝑠 =
invariant systems is the Laplace 𝐷(𝑠)
transform. The Laplace transform can be
written as a ratio of polynomials for a
large class of signals.

Source: pixabay

27
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 3 : Laplace Transform and Continuous -time LTI Systems

The Laplace transform, named after its Oliver Heaviside, an English electrical
inventor Pierre-Simon Laplace, is an integral engineer, suggested a comparable
transform in mathematics. It converts a technique without employing the
real-valued function t (typically time) into a Laplace transform and this results in an
complex-valued function s (complex operational calculus is known as the
frequency). Heaviside calculus.

In science and engineering, the transform Example:

has numerous uses. The Fourier transform
and the Laplace transform are comparable. Find the Laplace Transform and the ROC
A Fourier transform is a complex function of of X(t) = 𝑒 −𝑎𝑡 𝑢 𝑡 + 𝑒 𝑎𝑡 𝑢 −𝑡
a real variable (frequency), whereas a
Laplace transform is a complex function of a Solution :
complex variable.
From the Laplace Transform pairs table
In engineering and physics, the Laplace
transform is used to calculate the output of x(t) X(s) ROC
a linear time-invariant system by convolving
its unit impulse response with the input 𝑒 −𝑎𝑡 𝑢 𝑡 1 Re{s} > -a
signal. When this calculation is done in 𝑠+𝑎
Laplace space, the convolution becomes a −𝑒 𝑎𝑡 𝑢 −𝑡 1 Re{s} < -a
multiplication, which is easier to solve due 𝑠+𝑎
to its algebraic nature.

The Laplace transform, which is widely 1

𝐿 {𝑒 −𝑎𝑡 𝑢 𝑡 } = ; Re{s} > -a
𝑠+𝑎
utilised in mechanical and electrical
engineering, can also be used to solve
differential equations. The Laplace 1
transform converts a linear differential 𝐿{𝑒 𝑎𝑡 𝑢 −𝑡 } = - ; Re{s} < a
𝑠−𝑎
equation into an algebraic equation that can
be solved using algebraic techniques. The
inverse Laplace transform can then be used
to solve the original differential equation.

28
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 3 : Laplace Transform and Continuous -time LTI Systems

Therefore, 1
b) X(s) = , Re{s} < -1
𝑠+1
1 1
X(s) = - -a< Re{s} < a 1
𝑠+𝑎 𝑠−𝑎 c) X(s) = 2
𝑠 +4

(s−a) –(s+a) 1 1 2
= d) X(s) = - -
(𝑠+𝑎)(𝑠−𝑎) 𝑠+2 𝑠+5 𝑠+5

−2a 6
= e) X(s) = 2
(𝑠+𝑎)(𝑠−𝑎) 𝑠 +36

Thus, Solution:

Zeros : none By referring to the Laplace Transform pairs

Poles : s = -a , s=a table

1
The pole-zero diagram and ROC : a) x(t) = L-1 = 𝒆−𝒕 𝒖 𝒕
𝑠+1
Im
1
b) x(t) = L-1 = -𝒆−𝒕 𝒖 −𝒕
𝑠+1

1
c) x(t) = L-1 2
𝑠 +4
Re {s}
1
-𝑎 𝑎 = L-1
𝑠 +22
2

= cos 2t u(t)

1 1 2
d) x(t) = L-1 - L-1 - L-1
𝑠+2 𝑠+5 𝑠+5

Example: = 𝒆−𝟐𝒕 𝒖 𝒕 - 𝒆−𝟓𝒕 𝒖 𝒕 - 2𝒆−𝟓𝒕 𝒖 𝒕

Find the inverse Laplace transform of the 6

e) x(t) = L-1
following X(s) 2
𝑠 +36

1 6
a) X(s) = , Re{s} > -1 = L-1
𝑠+1 𝑠 +62
2

= sin 6t u(t)

29
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 3 : Laplace Transform and Continuous -time LTI Systems

Simulation
GNU Octave is a high-level programming language that is mostly used for numerical
calculations. Octave is a numerical solver for linear and nonlinear problems, as well as a
language for executing various numerical experiments that is mainly compatible with
MATLAB. Other characteristics of this software are as follows:
• Free software that runs on GNU/Linux, macOS, BSD, and Microsoft Windows
with a powerful mathematics-oriented syntax and built-in 2D/3D charting and
visualisation features.
• Compatible with a wide range of Matlab scripts.

GNU Octave is used to compute Laplace transform using the laplace function. The following
example shows the program arrays. It is important for the user to download and install the
recent symbolic bundle package from https://github.com/cbm755/octsympy/releases

Laplace Transform

For each of the given continuous-time signal, Laplace transform are computed using the
following command array

𝑥 𝑡 = 4 sin 100𝑡 𝑢(𝑡)

Program : Output :

syms t s X = (sym)
x=4*sin(100*t)*heaviside(t)
X=laplace(x,t,s) 400
----------
s2 + 10000

30
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 3 : Laplace Transform and Continuous -time LTI Systems

Alternatively, a simpler programing is also applicable to perform Laplace transform as

shown in below example.

x(t) = 3t3 + 4e5t

Program : Output :

syms s t
laplace ((3*t^3)+(4*e^(5*t)))

Inverse Laplace Transforms

GNU Octave can also compute the inverse Laplace Transforms using function ilaplace as
shown in the following example.

10(𝑠 + 1)
𝑋 𝑠 =
𝑠 2 + 4𝑠 + 3

Program : Output :

syms t s
x=10*(s+1)/(s^2+4*s+3)
ilaplace(x)
simplify(ans)
pretty(ans)

31
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 3 : Laplace Transform and Continuous -time LTI Systems

Poles and Zeros

The graphical representation of X(s) through its poles and zeros in the s-plane is referred to
as the pole-zero plot of X(s). GNU Octave can compute the poles and zeros of any given
continuous-time function and map the pole-zero plot as illustrated in the following
example.

10(𝑠 + 1)
𝑋 𝑠 =
𝑠 2 + 4𝑠 + 3

Program :

num=[10 10]
den=[1 4 3]
sys=tf(num, den)
pzmap(sys)
p,z]=pzmap(sys)

The GNU Octave mapped the output as illustrated in Fig. 3.1 with zeros at s=0 and two poles
at s=-1 and s=-3

Fig. 3.1: pole-zero plot

32
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 3 : Laplace Transform and Continuous -time LTI Systems

Source: pixabay

Exercises
1. Compute the Laplace transform of the following functions. Write the program in GNU
Octave using laplace command.

a) x(t) = e (-1/4) t
b) x(t) = 8 cos (10t) . 5u(2t)
c) x(t) = 2 cos 2t + 3 sin 2t
d) x(t) = e-t sin 5t + e-4t cos 3t

2. Using GNU Octave, compute the inverse Laplace Transforms of the following
functions.
10(s+1)
a) X s =
s2 +4s+8

6 2s 4
b) X s = + −
s s2 +16 s2

c) x(t) = 2 cos 2t + 3 sin 2t

d) x(t) = e-t sin 5t + e-4t cos 3t

33
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 3 : Laplace Transform and Continuous -time LTI Systems

3. Using GNU Octave, compute the poles and zeros of the following function.
10(𝑠+1)
a) X 𝑠 =
𝑠 2 +4𝑠+8

2s+100
b) X s =
s+1 s+8 (s+10)

34
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 4 : Z-transform and Discrete-time LTI Systems

Z-transform and
Discrete-time LTI
Systems
The Z-transform translates a discrete-time The inverse Z-transform is the reverse of
signal, which is a sequence of real or the Z-transform.
complex values, into a complex
frequency-domain representation in
mathematics and signal processing.

It is possible to specify the Z-transform as where C is a counterclockwise closed path

a one-sided or two-sided transform. The encircling the origin and entirely in the
formal power series X[z] defines the region of convergence (ROC). In the case
bilateral or two-sided Z-transform of a where the ROC is causal, this means the
discrete-time signal x[n]. path C must encircle all of the poles of
X[z].

𝑋(𝑧) = ෍ 𝑥[𝑛]𝑧 −𝑛
𝑛=−∞

Source: pixabay

35
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 4 : Z-transform and Discrete-time LTI Systems

The graphical representation of X(z) through 1

Zero : z = 0 , z=
12
its poles and zeros in the z-plane is referred
to as the pole-zero plot of X(z) 1 1
Poles : z= , z= -
2 3
Example:
The pole-zero plot of X(z) :
Find the Z-Transform and the ROC of x[n] = Im
1 1
( )𝑛 𝑢[𝑛] + (− )𝑛 𝑢[𝑛]
2 3

Solution :

From the Z-Transform pairs table

Re {z}
1 1 1
x[n] X(z) ROC - 0
3 12 2

1 𝑧 1
( )𝑛 𝑢[𝑛 |z| > | |
1 2
2 𝑧−
2
1 𝑧 1
(− )𝑛 𝑢[𝑛] |z| > | |
1 3
3 𝑧+
3
1 𝑧 1
𝑍 {( )𝑛 𝑢[𝑛]} = 1 ; |z| > | |
2 𝑧−2 2

1 𝑧 1
𝑍{(− )𝑛 𝑢[𝑛]} = 1 ; |z| > | |
3 𝑧+3 3

Therefore,
𝑧 𝑧
𝑋(𝑧) = +
1 1
𝑧− 𝑧+
2 3

2 z ( z − 121 )
=
( z − 12 )( z + 13 )

36
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 4 : Z-transform and Discrete-time LTI Systems

Simulation
WolframAlpha is a Wolfram Research computational knowledge engine and answer engine.
It directly responds to factual enquiries by computing the answer using data from external
sources.

The simulation for z-transform and inverse z-transform utilizes WolframAlpha online and
GNU Octave . The WolframAlpha computes z-transform . From the result, the GNU Octave is
used to compute the poles and zeros and map the pole and zeros plot.

Z-Transform

The WolframAlpha online widget for z-transform can be accessed online at the
WolframAlpha website as shown in Fig. 4.1.

Fig. 4.1 WolframAlpha online widget for z-transform

In this simulation, we use the sequence

1 1
x[n] = ( )𝑛 𝑢[𝑛] + ( )𝑛 𝑢[𝑛]
2 3

37
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 4 : Z-transform and Discrete-time LTI Systems

Key in the given sequence in the widget by using the heaviside command as follows

(((1/2)^n)*heaviside(n)) + (((1/3)^n)*heaviside(n))

Fig. 4.2 shows the z-transform of the given sequence,

Fig. 4.2 The z-transform output

The GNU Octave code commands for the pole and zeros plot are as follows. Although there
are two poles dan two zeros for the X(z), note that due to the programming contains, user
can only plot one pole and zero at a time. The pole and zeros plot are shown in Fig. 4.3. User
may use the same code command to plot the other pole and zero.

Program :

% zero and pole

b=[0] %b=zeros
a=[1/3] %a=poles
zplane(b,a)
title ("Zero pole plot");

38
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 4 : Z-transform and Discrete-time LTI Systems

Fig.12 The z-transform output

Fig.4.3 The pole and zero plot

Inverse z-Transform

The WolframAlpha online widget for inverse z-transform can be accessed online at
WolframAlpha website as shown in Fig. 4.4.

Fig. 4.4. WolframAlpha online widget for inverse z-transform

39
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 4 : Z-transform and Discrete-time LTI Systems

In this simulation, we use the sequence

10𝑧(𝑧+5)
X[𝑧] =
(𝑧−1)(𝑧−2)(𝑧+3)

Key in the given sequence in the widget as follows

(10z(z+5))/((z-1)(z-2)(z+3))

The WolframAlpha displays the inverse z-transform as shown in Fig. 4.5.

Fig. 4.5. The inverse z-transform output

40
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 4 : Z-transform and Discrete-time LTI Systems

Source: pixabay

Exercises
1. Using WolframAlpha Online, compute the z-transform of the following functions.
Then compute the poles and zeros from the result using Octave Online. Write the
program.
1 1
a) x[n] = ( )𝑛 𝑢[𝑛] + ( )𝑛 𝑢[−𝑛 − 1]
3 2

1 1
b) x[n] = ( )𝑛 𝑢[𝑛] + ( )𝑛 𝑢[−𝑛 − 1]
2 3

2. Using WolframAlpha Online, compute the inverse z-transform of the following

functions.
𝑧2
a) X[𝑧] =
(2z+1)(z−1)

4(2𝑧+1)
b) X[𝑧] =
(𝑧+1)(𝑧−3)

41
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 5 : Fourier Analysis Of Continuous -time Signal and System

Fourier Analysis Of
Continuous-time Signal
and System
The Laplace transform and the z- Fourier analysis is necessary for
transform have been introduced in the describing certain types of systems and
previous chapters as to convert time- their attributes in the frequency domain,
domain signals into complex s-domain in addition to generating spectral
and z-domain. This representations are representations of signals. We will cover
more straightforward to examine and Fourier analysis in the context of
process for various reasons. These continuous-time signals and systems in
modifications also reveal more this chapter.
information about the nature and
behavior of various signals and systems.

We will introduce further transformations

such as Fourier series and Fourier
transform that convert time-domain
signals into frequency domain (or
spectral) representations in this and the
subsequent chapters.

Source: pixabay

42
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 5 : Fourier Analysis Of Continuous -time Signal and System

Trigonometric Fourier Series

The trigonometric Fourier series Example:

representation of a periodic signal x(t)
Determine the trigonometric Fourier
with fundamental period T0 is given by
∞
Series of the following signal.
𝑎0
𝑓(𝑡) = + ෍ {𝑎𝑘 cos 𝑘 𝜔0 𝑡 + 𝑏𝑘 sin 𝑘 𝜔0 𝑡}
2
𝑘=1

where ak and bk are the Fourier

coefficients given by
𝑇/2
2
𝑎𝑘 = න 𝑓 𝑥 cos 𝑘 𝜔0 𝑡 𝑑𝑡 Fig. 5.1
𝑇0
−𝑇/2
Solution :
𝑇/2
2 Step 1 : Identify type of given function
𝑏𝑘 = න 𝑓 𝑥 sin 𝑘 𝜔0 𝑡 𝑑𝑡 either odd or even function
𝑇0
−𝑇/2
It is an odd function. Thus , ak = 0
If a periodic signal f(t) is even, then bk = 0
and its Fourier series contains only cosine The Fourier series is
terms ∞
∞
𝑎0 𝑓(𝑡) = ෍ 𝑏𝑘 sin 𝑘 𝜔0 𝑡
𝑓(𝑡) = + ෍ 𝑎𝑘 cos 𝑘 𝜔0 𝑡
2 𝑘=1
𝑘=1
Step 2 : Identify the fundamental period
If f(t) is odd, then ak = 0 and its Fourier 𝑇0 and 𝜔0
series contains only sine terms
∞ 𝑇0 = 2𝜋
𝑓(𝑡) = ෍ 𝑏𝑘 sin 𝑘 𝜔0 𝑡
𝑘=1 2𝜋 2𝜋
𝜔0 = = =1
𝑇0 2𝜋

43
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 5 : Fourier Analysis Of Continuous -time Signal and System

Step 3 : Find the Fourier coefficients

Thus, trigonometric Fourier Series is

∞
2
𝑓(𝑡) = ෍ 1 − (−1)𝑘 sin 𝑘 𝜔0 𝑡
𝑘𝜋
𝑘=1

44
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 5 : Fourier Analysis Of Continuous -time Signal and System

Simulation
User can learn how to utilize octave-online software to execute Fourier Transform of
Continuous-Time and find the Fourier coefficient

Trigonometric Fourier Series

We may utilize the heaviside(t) command in GNU Octave to plot a diagram of a periodic
continuous-time signal as shown in Fig. 9. The Fourier coefficient are found with the
following command array

𝑓(𝑡)

𝑡
0 1 2 3

Fig. 9 : Periodic Signal

%Fourier Transform Of Continuous-Time

syms t
f=heaviside(t)-heaviside(t-1)

%Diagram of continuous-time
ezplot(f,[0,2])
T=2

%Fourier coefficients a0=c0

a0_sym=1/T*int(f,t,0,T)
double(a0_sym)
w0=2*pi/T

45
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 5 : Fourier Analysis Of Continuous-time Signal and System

%Fourier coefficients
a1=2/T*int(f*cos(1*w0*t),t,0,T)
a2=2/T*int(f*cos(2*w0*t),t,0,T)
a3=2/T*int(f*cos(3*w0*t),t,0,T)

%Fourier coefficients
b1=2/T*int(f*sin(1*w0*t),t,0,T)
b2=2/T*int(f*sin(2*w0*t),t,0,T)
b3=2/T*int(f*sin(3*w0*t),t,0,T)

GNU Octave displays the Fourier coefficient a1,a2,a3,b1,b2 and b3 as in Table 1

Table 1 Fourier coefficient from GNU Octave

Fourier Coefficient Output

a1 a1 = (sym) 0

a2 a2 = (sym) 0

a3 a3 = (sym) 0
b1 = (sym)
2
b1
--
pi
b2 b2 = (sym) 0
b3 = (sym)
2
b3
----
3*pi

46
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 5 : Fourier Analysis Of Continuous -time Signal and System

Source: pixabay

Exercises
1. Execute the following integral function using GNU Octave as per example command
given.

47
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 6 : Fourier Analysis Of Discrete -time Signal and System

Fast Fourier Transform

The discrete Fourier transform (DFT) of a By factoring the DFT matrix into a product
sequence, or its inverse, is computed of sparse (mainly zero) elements, an FFT
using a fast Fourier transform (FFT) (IDFT). may quickly compute such modifications.
Fourier analysis transforms a signal from The speed difference can be huge,
its original domain (typically time or especially for large data sets with N in the
space) to a frequency domain hundreds or millions. Many FFT
representation and back. Decomposing a techniques are substantially more
sequence of values into components of accurate than directly or indirectly
different frequencies yields the DFT. This evaluating the DFT definition in the
procedure is useful in a variety of presence of round-off error. FFT
domains, but computing it straight from a algorithms are based on a variety of
definition is typically too time consuming published theories, ranging from simple
to be practical. complex-number arithmetic to group
theory and number theory.

By factoring the DFT matrix into a product

of sparse (mainly zero) elements, an FFT
may quickly compute such modifications.
The speed difference can be huge,
especially for large data sets with N in the
hundreds or millions. Many FFT
techniques are substantially more
accurate than directly or indirectly
evaluating the DFT definition in the
presence of round-off error. FFT
algorithms are based on a variety of
published theories, ranging from simple
complex-number arithmetic to group
theory and number theory.

48
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 6 : Fourier Analysis Of Discrete -time Signal and System

Simulation
Fast Fourier Transform

User can learn how to utilize GNU Octave to simulate Fourier analysis on discrete-time
signal and system. The simulation is to covert a noise corrupted signal y(t) in time domain
into frequency domain . First, a sound signal is created as shown in Fig. 6.1. Then, a random
noise signal is created as shown in Fig. 6.2. This noise is added in the sound signal producing
a corrupted signal as shown in Fig. 6.3. Next, fast Fourier Transform (FFT) is performed on
the signal to calculate and plot the power spectrum as shown in Fig. 6.4. The number of
sampling points is at 1000Hz for 5 seconds. This simulation is done using the following
command array .

clc
clear all
% To convert a noise corrupted signal y in time domain into a frequency domain
% number of sampling points at 1000Hz for 5 seconds
t2=0:0.001:0.5;
x2 = sin(2*pi*50*t2) + sin(2*pi*120*t2); % a sound signal
figure(1)
plot(1000*t2(1:50),x2(1:50))
title('A sound signal x2');xlabel('Time(milliseconds)')
y = x2 + 2*randn(size(t2)); % create random noise to be added to the sound signal.
figure(2)
plot(1000*t2(1:50),y(1:50)) % plot the corrupted sound signal
title('Signal Corrupted with zero-mean Random noise');xlabel('time(milliseconds)')
% will perform 1 512-points fast Fourier transform(FFT), calculate and plot the
power spectrum
Y = fft(y,512);
Pyy = Y.*conj(Y)/512;
f = 1000*(0:256)/512;
figure(3)
plot(f,Pyy(1:257))
title('Frequency content of y');xlabel('frequency (Hz)')

49
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 6 : Fourier Analysis Of Discrete -time Signal and System

GNU Octave displays discrete-time input x[n], impulse response h[n] and convolution
output y[n] as shown in Fig. 6.1 to Fig. 6.4.

Fig. 6.1 : A sound signal

Fig. 6.2 : Random noise signal

50
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 6 : Fourier Analysis Of Discrete -time Signal and System

Fig. 6.3 : Signal corrupted with zero-mean random noise

Fig. 6.4 : Frequency content of signal y(t)

51
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 Chapter 6 : Fourier Analysis Of Discrete-time Signal and System

Source: pixabay

Exercises
Consider the periodic square wave x(t) shown in Fig. 6.5. Find the Fourier Series for the
function for which the graph is given below

Fig. 6.5

52
SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
 REFERENCES
Hwei P.Hsu. (2014). Schaums Outline of Digital Signal Processing (4th ed).,
United States: McGraw-Hill Education.
Steven T.Karris (2009). Digital Signal Processing Using MATLAB & Wavelets.
2nd Edition. Jones and Bartlett Publishers, Inc. Andreas Antoniou.
Sylvia Ong, Dyg Khayrunsalihaty. (2021). Signal and System : Augmented-
Reality Experience. Politeknik Kuching Sarawak.

Icon
www.pixabay.com

SIG N AL A N D SYST EM W IT H G N U O C T A VE
SIMU L A T IO N
Fauziah Aliman is the senior lecturer at Politeknik Merlimau,
Melaka (PMM). She earned a Bachelor's Degree in Electronic
Engineering at Multimedia University (MMU), Melaka and
Master's Degree in Electronic Engineering (System Electronics) at
Universiti Teknikal Malaysia Melaka (UTeM) . She is a registered
graduate engineer by Board of Engineers Malaysia. She has over
seven years of engineering experience in semiconductor industry.
In addition, she has over fourteen years of teaching experience
that she acquired at Technical and Vocational Education and
Training (TVET) institutions. She is currently with Department of
Electrical Engineering of Politeknik Merlimau.

Syamsul Bahri Bin Mohamad is a Lecturer in the Department of

Electrical Engineering, Politeknik Merlimau, Melaka (PMM). He
holds a Master's Degree (Communication and Computer) from
Universiti Kebangsaan Malaysia and a Master's Degree (Technical
and Vocational Education) as well as a Bachelor's Degree
(Electrical Engineering) from Kolej Universiti Teknologi Tun Hussein
Onn (KUiTTHO) which is now Universiti Tun Hussein Onn (UTHM).
He is a lecturer for the Signal & System at Electrical Engineering
Department, Politeknik Merlimau, which is one of the courses
offered at the Malaysia polytechnic.

671