FOURIER SERIES

IS3302
Complex Analysis and Mathematical Transforms
Department of Interdisciplinary Studies
Faculty of Engineering
University of Ruhuna
Content
• Introduction to Signals and Mathematical Transforms
• Periodic Functions
• Harmonics, sinusoidal Functions and Non-Sinusoidal Functions
• Even and Odd Functions
• Representing Periodic Functions by Fourier Series
• Half – Range Series
• Fourier series for functions of general period
• Half-Range Cosine series for functions of general period
• Parseval’s Formula
• Complex form of Fourier Series
Essential Types of Signals in Signal Processing
• Continuous-time and discrete-time signals.

• Even and Odd Signals.

• Periodic and non-periodic signals.

• Deterministic and random signals.

• Energy and Power types Signals.

 Essential Types of Signals in Signal Processing
• Continuous-time
• A signal is considered to be a continuous time signal if it is defined over a continuum
of the independent variable (If a signal can take any value on the x-axis (time axis)
then it is called as continuous signal).

• discrete-time signals
• A signal is considered to be discrete time if the independent variable only has
discrete values (if it can only take finite values on x-axis (time axis) then it will be a
discrete signal).
Essential Types of Signals in Signal
Processing…
• Even and Odd Signals.
• Even signals are symmetric around vertical
axis, and Odd signals are symmetric about
origin.

• Even Signal: A signal is referred to as an even if

it is identical to its time-reversed counterparts;
x(t) = x(-t).

• Odd Signal: A signal is odd if x(t) = x(-t).

Essential Types of Signals in Signal Processing
• Periodic signals Non-periodic signals
• A periodic signal is one that repeats the sequence of • A non-periodic or aperiodic signal is one for which no value
values exactly after a fixed length of time, known as the of T satisfies Equation
period.
• 𝒙 𝒕 = 𝒙(𝒕 + 𝑻)
• 𝑥 𝑡 = 𝑥(𝑡 + 𝑇) • a non-periodic signal is a signal that does not repeat itself
1 at regular intervals.
• 𝑓 = ; f= number of cycles per second, T=Period
𝑇 • This means that it does not have a fundamental frequency,
and its frequency spectrum is generally not composed of
discrete frequencies.
• Periodic signals are characterized by their period, which
is the time it takes for the signal to repeat itself, and their • Non-periodic signals can be continuous or discrete, and
frequency, which is the number of cycles the signal goes they can be either deterministic or random.
through in a given time period. • Examples of non-periodic signals include white noise,
• impulses, and arbitrary waveforms.
• The period and the frequency of a periodic signal are • Non-periodic signals are often encountered in real-world
inversely related: a signal with a short period has a high applications, such as speech, music, and environmental
frequency, and a signal with a long period has a low noise.
frequency.
Essential Types of Signals in Signal Processing
• Deterministic signals
• A signal is said to be deterministic if there is no
uncertainty with respect to its value at any instant of
time.
• Example, 𝑥 𝑡 = sin(3𝑡)

• Random signals (Non-deterministic signals )

• Non-deterministic signals are random in nature
hence they are called random signals.
• Random signals cannot be described by a
mathematical equation.
noise
Essential Types of Signals in Signal Processing

• Energy types Signals.

• Signals that are both deterministic and non-periodic are energy signals.

• Power types Signals

• Periodic and random signals are power signals.
• Mathematical transforms are essential tools in many scientific and engineering applications and are crucial in
various fields, including signal processing, image recognition, and data analysis.
• They allow for representing complex signals and images in a more efficient and manageable form.

• By applying transforms,
• it can extract valuable information,
• reduce noise, and
• enhance the quality of the signal.

• Transforms enable you to work with signals in a more abstract and manipulable way, making it easier to
• identify patterns,
• analyze data, and
• make predictions.

• Mathematical transforms are techniques used to represent functions or signals in a different domain.

• Simply,
• Fourier series/Fourier Transform: decomposes a function into its frequency components

• Laplace Transform: used for continuous-time signals, transforms a function from the time domain to the s-domain

• Z-Transform: used for discrete-time signals and has applications in signal processing and control systems
 FOURIER SERIES
• The Fourier series is a fundamental
concept in mathematics and signal
processing!

• It's a way to represent a periodic

function as a sum of simpler sine and
cosine waves.

• It's used to analyze and decompose

signals, understand periodic
phenomena, and solve partial
differential equations.

• The concept has numerous practical

uses, such as filtering noise,
encrypting data, and even generating
music and images.
 FOURIER TRANSFORMS
• Fourier transforms decomposes signals or functions into their constituent frequencies,
providing valuable insights into the underlying structure and behavior.

• Fourier transforms are essential in many applications, such as

• filtering,
• modulation, and demodulation,
• as well as in analyzing and understanding natural phenomena like sound, light, and vibrations.
 LAPLACE TRANSFORM
• The Laplace transform is a fundamental tool in the field of mathematics and electrical
engineering.
• It transforms a continuous-time signal into a frequency domain representation, which
makes it easier to analyze and manipulate the signal.
• The Laplace transform is used to solve differential equations, determine the stability of
systems, and analyze the frequency response of Linear Time-Invariant (LTI) systems.
• It is also used in many real-world applications, such as signal processing, control systems,
and communications.
 Z-TRANSFORM
• The concept of Z-transform is a fundamental aspect of signal processing and control systems.

• The Z-transform is a discrete-time variant of the Laplace transform, used to analyze and design digital control systems.

• It transforms a continuous-time signal into a discrete-time representation, making it easier to analyze and process.

• The Z-transform is used to represent the frequency response of a system, allowing engineers to design and optimize
their systems for specific frequency responses.

• It's a crucial tool in many fields, including control systems, signal processing, and digital filter design
 Introduction
• Fourier series are usually infinite series but involve sine and cosine
functions (or their complex exponential equivalents) rather than
polynomials.

• Fourier series are used for approximating periodic functions/in the analysis
of periodic functions.

• Many of the phenomena studied in engineering and

science are periodic in nature.

• The sum of these special trigonometric functions is called the Fourier

Series.
 Periodic Functions
A 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝑓(𝑡) is said to be periodic with period 𝑝 if

𝑓(𝑡 + 𝑝) = 𝑓(𝑡)
for all values of 𝑡 and if 𝑝 > 0.

• The period of the function 𝑓(𝑡) is the interval between two successive
repetitions.

• Periodic functions have repetitive behaviour.

 Periodic Functions…
• The most obvious examples of periodic functions are the trigonometric
functions sin 𝑡 and cos 𝑡, both of which have period 2𝜋 (using radian
measure)
Periodic Functions…
 Periodic Functions…
• The amplitude of these sinusoidal functions is the maximum displacement from y = 0
and is clearly 1.
• More generally we can consider a sinusoid 𝑦 = 𝐴𝑠𝑖𝑛 𝑛𝑡 which has maximum value,
or amplitude, 𝐴 and where 𝑛 is usually a positive integer.

Example
2𝜋
𝑦 = sin 2𝑡 is a sinusoid of amplitude 1 and period = 𝜋.
2
The fact that the period is 𝜋 follows because
sin 2(t + 𝜋 ) = sin(2t + 2 𝜋) = sin 2t
for any value of t.
 Periodic Functions… 2𝜋
In general 𝑦 = 𝐴𝑠𝑖𝑛 𝑛𝑡 has amplitude 𝐴, period and completes 𝑛 oscillations when 𝑡
𝑛
changes by 2.

Formally, we define the frequency of a sinusoid as the reciprocal of the period:

1
𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦(𝑓) = 𝑓 = 1/𝑇
𝑃𝑒𝑟𝑖𝑜𝑑

and the angular frequency, often denoted the Greek Letter 𝜔 (omega) as
2𝜋
𝑎𝑛𝑔𝑢𝑙𝑎𝑟 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦(𝜔) = 2𝜋 × 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 =
𝑝𝑒𝑟𝑖𝑜𝑑
𝜔 = 2𝜋𝑓
𝒏
Thus 𝑦 = 𝐴𝑠𝑖𝑛 𝑛𝑡 has frequency and angular frequency 𝒏.
𝟐𝝅

State the amplitude, period, frequency and angular frequency of

2𝑡
1. 𝑦 = 5𝑐𝑜𝑠4𝑡 2. 6𝑠𝑖𝑛
3
 Sinusoids (A Sine Wave)
A curve similar to the sine function but possibly shifted in
phase, period, amplitude, or any combination thereof.

i.e. if the function can be produced by stretching, shifting or

compressing of sine function (y= 𝐴𝑠𝑖𝑛(𝑛𝑡)).

The general sinusoid of amplitude 𝐴, angular frequency 𝜔,

2𝜋
(and period ), and phase 𝑐 is given by cosine is a sinusoidal
𝜔
function. You can think of it
as the sine function with a
𝑓 𝑥 = 𝐴𝑠𝑖𝑛(𝜔𝑥 + 𝑐) phase shift of -pi/2 (or a
phase shift of 3pi/2).
 Examples
𝑦 = 2𝑠𝑖𝑛𝑥 ; 𝑥 = 0 𝑡𝑜 𝑥 = 2𝜋
𝑦 = sin(𝑥 − 𝜋) ; 𝑥 = 0 𝑡𝑜 𝑥 = 2𝜋

𝑦 = 𝑠𝑖𝑛2𝑥 ; 𝑥 = 0 𝑡𝑜 𝑥 = 2𝜋
𝑦 = 1 + 𝑠𝑖𝑛𝑥 ; 𝑥 = 0 𝑡𝑜 𝑥 = 2𝜋
Harmonics
Non-sinusoidal periodic functions
Non-sinusoidal periodic functions…
Non-sinusoidal periodic functions…
Non-sinusoidal periodic functions…

• Write down an analytic definition for the following periodic function:

Even and Odd Functions
• A function 𝑓(𝑥) is said to be even if 𝑓 −𝑥 = 𝑓 𝑥 .
• The function 𝑓(𝑥) is said to be odd if 𝑓 −𝑥 = −𝑓 𝑥 .

• Graphically, even functions have symmetry about the y-axis, whereas

odd functions have symmetry around the origin.
 Even and Odd Functions…
Examples

• Sums of odd powers of x are odd: 5𝑥 3 − 3𝑥

• Sums of even powers of x are even: −𝑥 6 + 4𝑥 4 + 𝑥 2 − 3
• sin 𝑥 is odd, and cos 𝑥 is even

• The product of two odd functions is even: 𝑥 sin 𝑥 is even

• The product of two even functions is even: 𝑥 2 cos 𝑥 is even
• The product of an even function and an odd function is odd: sin 𝑥 cos 𝑥 is odd
Even and Odd Functions…
Even and Odd Functions…
Representing Periodic Functions by Fourier Series
Here we discuss how a periodic function can be expressed as a series of
sines and cosines.

If 𝑓(𝑡) is a periodic function, of period 2𝜋, then the Fourier series

expansion takes the form:
∞
𝑎0
𝑓 𝑡 = + (𝑎𝑛 cos 𝑛𝑡 + 𝑏𝑛 sin 𝑛𝑡)
2
𝑛=1

Here for 𝑎𝑛 ; 𝑛 = 0,1,2,3, … and 𝑏𝑛 ; 𝑛 = 1,2,3, … are Fourier

Coifficients.
 Some useful Integrals
2𝜋 2𝜋
0
sin 𝑛𝑥 𝑑𝑥 =0 0
cos 𝑛𝑥 𝑑𝑥 =0

2𝜋 2 𝑛𝑥𝑑𝑥 2𝜋 2 𝑛𝑥𝑑𝑥
0
𝑠𝑖𝑛 =𝜋 0
𝑐𝑜𝑠 =𝜋

2𝜋 2𝜋
0
sin 𝑛𝑥 sin 𝑚𝑥 𝑑𝑥 =0 0
cos 𝑛𝑥 cos 𝑚𝑥 𝑑𝑥 =0

2𝜋 2𝜋
0
sin 𝑛𝑥 cos 𝑚𝑥 𝑑𝑥 =0 0
sin 𝑛𝑥 cos 𝑛𝑥 𝑑𝑥 =0

𝑛
sin 𝑛𝜋 = 0, cos 𝑛𝜋 = −1 ; 𝑛∈𝐼
 Calculation of Fourier coefficients
Consider the Fourier Series for a function 𝑓(𝑡) of period 2𝜋 ;
∞
𝑎0
𝑓 𝑡 = + (𝑎𝑛 cos 𝑛𝑡 + 𝑏𝑛 sin 𝑛𝑡)
2
𝑛=1

To obtain the coefficients 𝑎𝑛 (𝑛 = 1, 2, 3, . . . ), multiply both sides by

𝑐𝑜𝑠𝑚𝑡 where 𝑚 is some positive integer and Integrate both sides from
− 𝜋 to 𝜋 (or 0 to 2𝜋).
𝜋 𝜋 ∞ 𝜋 𝜋
𝑎0
f(t)cos 𝑚𝑡 𝑑𝑡 = cos 𝑚𝑡 𝑑𝑡 + 𝑎𝑛 cos 𝑛𝑡 cos 𝑚𝑡 𝑑𝑡 + 𝑏𝑛 sin 𝑛𝑡 cos 𝑚𝑡 𝑑𝑡
2
−𝜋 −𝜋 𝑛=1 −𝜋 −𝜋
Calculation of Fourier coefficients…
𝜋 𝜋

f(t)cos 𝑚𝑡 𝑑𝑡 = 𝑎𝑚 𝜋 ( 𝑐𝑜𝑠 2 𝑚𝑡𝑑𝑡 = 𝜋)

−𝜋 −𝜋

Rewriting 𝑚 as 𝑛,
𝜋
1
𝑎𝑛 = f(t)cos 𝑛𝑡 𝑑𝑡 ; 𝑛 = 1,2,3, …
𝜋
−𝜋
 Calculation of Fourier coefficients…
By multiplying the first equation by sin 𝑚𝑡, and integration,
𝜋 𝜋 ∞ 𝜋 𝜋
𝑎0
f t sin 𝑚𝑡 𝑑𝑡 = sin 𝑚𝑡 𝑑𝑡 + 𝑎𝑛 cos 𝑛𝑡 sin 𝑚𝑡 𝑑𝑡 + 𝑏𝑛 sin 𝑛𝑡 sin 𝑚𝑡 𝑑𝑡
2
−𝜋 −𝜋 𝑛=1 −𝜋 −𝜋

All the terms on the R.H.S equal to zero except the case 𝑛 = 𝑚.
𝜋 𝜋

f t sin 𝑚𝑡 𝑑𝑡 = 𝑏𝑚 𝜋 ( 𝑠𝑖𝑛2 𝑚𝑡𝑑𝑡 = 𝜋)

−𝜋 −𝜋
Calculation of Fourier coefficients…
𝜋
1
𝑏𝑛 = f t 𝑠𝑖𝑛 𝑚𝑡 𝑑𝑡
𝜋
−𝜋

Rewriting 𝑚 as 𝑛,
𝜋
1
𝑏𝑛 = f t 𝑠𝑖𝑛 𝑛𝑡 𝑑𝑡 ; 𝑛 = 1,2,3, …
𝜋
−𝜋
 Remarks
A function 𝑓(𝑡) with period 2𝜋 has a Fourier Series,
∞
𝑎0
𝑓 𝑡 = + (𝑎𝑛 cos 𝑛𝑡 + 𝑏𝑛 sin 𝑛𝑡)
2
𝑛=1

The Fourier Coefficients are,𝜋

1
𝑎𝑛 = f(t)cos 𝑛𝑡 𝑑𝑡 ; 𝑛 = 0,1,2,3, …
𝜋
−𝜋𝜋
1
𝑏𝑛 = f t 𝑠𝑖𝑛 𝑛𝑡 𝑑𝑡 ; 𝑛 = 1,2,3, …
𝜋
−𝜋
Example
• Obtain the Fourier Series of the function 𝑓(𝑡) which is shown below.

1 ;0 < 𝑡 < 𝜋
𝑓 𝑡 =
0 ; 𝜋 < 𝑡 < 2𝜋
Example
• Obtain the Fourier Series of the square wave one period of which is
shown,
Expansion of Even Function
∞
𝑎0
𝑓 𝑡 = + (𝑎𝑛 cos 𝑛𝑡)
2
𝑛=1
The Fourier Coefficients are, 𝜋
2
𝑎0 = 𝑓 𝑡 𝑑𝑡
𝜋
0

𝜋
2
𝑎𝑛 = f(t)cos 𝑛𝑡 𝑑𝑡 ; 𝑛 = 0,1,2,3, …
𝜋
0
𝑏𝑛 = 0
Expansion of Odd Function
∞

𝑓 𝑡 = (𝑏𝑛 sin 𝑛𝑡)

𝑛=1
The Fourier Coefficients are,
𝑎0 = 𝑎𝑛 = 0
𝜋
2
𝑏𝑛 = f t sin 𝑛𝑡 𝑑𝑡 ; 𝑛 = 0,1,2,3, …
𝜋
0
Examples
• Find the Fourier series expansion of the periodic function of period
2𝜋,
𝑓 𝑥 = 𝑥 2 ; −𝜋 ≤ 𝑥 ≤ 𝜋

• Obtain the Fourier series expression for,

𝑓 𝑥 = 𝑥 3 ; −𝜋 ≤ 𝑥 ≤ 𝜋
Half – Range Series, Period 0 𝑡𝑜 𝜋
• To get the series of Cosines only we assume that the function 𝑓(𝑥) is an even function in the
interval (−𝜋, 𝜋).
Then
∞
𝑎0
𝑓 𝑡 = + (𝑎𝑛 cos 𝑛𝑡)
2
𝑛=1
The Fourier Coefficients are, 𝜋
2
𝑎0 = 𝑓 𝑡 𝑑𝑡
𝜋
0

𝜋
2
𝑎𝑛 = f(t)cos 𝑛𝑡 𝑑𝑡 ; 𝑛 = 0,1,2,3, …
𝜋
0
𝑏𝑛 = 0
Half – Range Series, Period 0 𝑡𝑜 𝜋
• To get the series of Sines only we assume that the function 𝑓(𝑥) is an odd
function in the interval (−𝜋, 𝜋).
Then
∞

𝑓 𝑡 = (𝑏𝑛 sin 𝑛𝑡)

𝑛=1
The Fourier Coefficients are,
𝑎0 = 𝑎𝑛 = 0
𝜋
2
𝑏𝑛 = f t sin 𝑛𝑡 𝑑𝑡 ; 𝑛 = 0,1,2,3, …
𝜋
0
Example
• Represent the following function by a Fourier sin series.

𝜋
𝑡; 0<𝑡≤
𝑓 𝑡 = 𝜋 𝜋 2
; <𝑡≤𝜋
2 2
 Fourier series for functions of general period
• a 2𝜋 periodic function 𝑓(𝑥) has a Fourier series

• With

𝑧𝑐
• Suppose we now change the variable to 𝑧 where 𝑥 =
𝜋
• Thus 𝑥 = 𝜋 corresponds to 𝑧 = 𝑐 and 𝑥 = −𝜋 corresponds to 𝑧 = −𝑐
• Hence regarded as a function of 𝑧, we have a function with period 2𝑐.
 𝜋𝑥 𝑧𝑐
Making the substitution 𝑧 = or (𝑥 = )
𝑐 𝜋
𝜋
and hence dz = 𝑑𝑥
𝑐
Then the function 𝑓 𝑥 of period 2𝑐 is transformed to the function
𝑐𝑧
𝑓 = 𝐹(𝑧) of period 2𝜋.
𝜋
Then
2𝑐 2𝑐 2𝑐
1 1 𝑛𝜋𝑥 1 𝑛𝜋𝑥
𝑎0 = 𝑓 𝑥 𝑑𝑥 , 𝑎𝑛 = f(x)cos 𝑑𝑥, 𝑏𝑛 = f x sin 𝑑𝑥
𝑐 𝑐 𝑐 𝑐 𝑐
0 0 0
Half-Range Cosine series for functions of general period

∞
𝑎0 𝑛𝜋𝑥
𝑓 𝑥 = + (𝑎𝑛 cos )
2 𝑐
𝑛=1
The Fourier Coefficients are, 𝑐
2
𝑎0 = 𝑓 𝑥 𝑑𝑥
𝑐
0
𝑐
2 𝑛𝜋𝑥
𝑎𝑛 = f(x)cos 𝑑𝑥 ; 𝑛 = 0,1,2,3, …
𝑐 𝑐
0
𝑏𝑛 = 0
Half-Range Sine series for functions of general period

∞
𝑛𝜋𝑥
𝑓 𝑥 = (𝑏𝑛 sin )
𝑐
𝑛=1
The Fourier Coefficients are,
𝑎0 = 𝑎𝑛 = 0
𝑐
2 𝑛𝜋𝑥
𝑏𝑛 = f x sin 𝑑𝑥 ; 𝑛 = 0,1,2,3, …
𝜋 𝑐
0
Example
• Find the Fourier series expansion of 𝑓 𝑥 .
𝑓 𝑥 = 𝑥 , −2 < 𝑥 < 2.
Parseval’s Formula
𝑐 ∞
2 𝑑𝑥
1 2
𝑓(𝑥) = 𝑐 𝑎0 + 𝑎𝑛2 + 𝑏𝑛2
2
−𝑐 𝑛=1
We know that
∞
𝑎0 𝑛𝜋𝑥 𝑛𝜋𝑥
𝑓 𝑥 = + (𝑎𝑛 cos + 𝑏𝑛 sin )
2 𝑐 𝑐
𝑛=1
Multiplying by 𝑓(𝑥) we have
∞
2
𝑎0 𝑛𝜋𝑥 𝑛𝜋𝑥
𝑓(𝑥) = 𝑓 𝑥 + (𝑎𝑛 𝑓 𝑥 cos + 𝑏𝑛 𝑓 𝑥 sin )
2 𝑐 𝑐
𝑛=1
Parseval’s Formula…
• Integrating
𝑐
term by term
𝑐
from −𝑐 𝑡𝑜 𝑐, we 𝑐have 𝑐
∞ ∞
2
𝑎0 𝑛𝜋𝑥 𝑛𝜋𝑥
𝑓(𝑥) 𝑑𝑥 = 𝑓 𝑥 𝑑𝑥 + (𝑎𝑛 𝑓 𝑥 cos + (𝑏𝑛 𝑓 𝑥 sin 𝑑𝑥
2 𝑐 𝑐
−𝑐 −𝑐 𝑛=1 −𝑐 𝑛=1 −𝑐
𝑐 ∞
2 𝑑𝑥
𝑎0
𝑓(𝑥) = 𝑎0 𝑐 + (𝑎𝑛 𝑐𝑎𝑛 + 𝑏𝑛 𝑐𝑏𝑛 )
2
−𝑐 𝑐 𝑛=1
2 ∞
𝑎 0
𝑓(𝑥) 2 𝑑𝑥 = 𝑐 + (𝑎𝑛2 𝑐 + 𝑏𝑛2 𝑐)
2
−𝑐 ∞ 𝑛=1
𝑎02
=𝑐 + (𝑎𝑛2 + 𝑏𝑛2 )
2
𝑛=1
Example
Answer
Example
Example
Example
Example
Example
Example
• Find the Fourier Series of the triangular signal shown below.

• Then use Parseval’s Theorem to show that

Answer
 Complex form of Fourier Series
• Let the function f (x) be defined on the interval [−π, π]. Using the well-
known Euler's formulas
𝑒 𝑖𝑥 +𝑒 −𝑖𝑥 𝑒 𝑖𝑥 −𝑒 −𝑖𝑥
• cos 𝑥 = sin 𝑥 =
2 2𝑖
• we can write the Fourier series of the function in complex form:

• Here we have used the following notations:

• The coefficients cn are called complex Fourier coefficients. They are defined
by the formulas

• If necessary to expand a function f(x) of period 2L, we can use the following
expressions:

• where
Example 1.
Using complex form, find the Fourier series of the function