Unit 03

Continuous Time Fourier Series and Transform

• Unit 3: Continuous Time Fourier series and Transform
• Fourier Series
• Introduction of Fourier series
• Fourier Integral
• Representation of aperiodic and periodic signals
• Fourier Transform
• Forward and reverse/inverse Fourier transforms
• Fourier transforms properties.
• Dirichlet condition, Parseval's theorems

Text: Alan V. Oppenheim, Alan S. Willsky with S. Hamid Nawab, Signals and Systems, Prentice Hall, 1996
Reference : Fawwaz T. Ulaby and Andrew E. Yagle, Signals and Systems: Theory and Applications, Michigan Publishing, 2018
L-13 and L-14

FOURIER SERIES
Motivation
• Because of the superposition property, the response of an LTI system to any
input consisting of a linear combination of basic signals is the same linear
combination of the individual responses to each of the basic signals.
– Representation of signals as linear combinations of a set of basic signals.
• Representation of signal and analysis of signal properties
– Representation of system for computation of response to inputs based on system
response to basic signal input.
• Representation of system and analysis of system response
• Impulse Response and Convolution
– Representing signals as linear combinations of shifted impulses
– Representation of system using impulse response.
– Representation and analysis of LTI systems through the convolution sum and integral
• Fourier Series and Transform
– Representation of signal using sinusoids, phasors, and complex exponentials
– Representation of system using impedance and frequency response
Existence of Periodic Sinusoidal Signals in Nature
• Damped and under-damped sinusoids
describe the variables that exist in many
natural or physical systems.
– Electric power generation, transmission, and
distribution
– Communications using carrier waves
– Vibrations and waves

Waves in the ocean consist of the linear combination of sinusoidal String vibration
waves with different spatial periods or wavelengths
Composite Sinusoid
Time and Phasor Domain Representation of Sinusoid

We want to go from time to phasor (amplitude, phase, lead, lag)

to frequency and transform domains
Response of Electric Circuits to Periodic Signal
Fourier Series Representation of Periodic Signals

• The {an, bn, cn, xn} are called

Fourier coefficients.
Fourier Series Interpretation
• A Fourier series is a mathematical version of a prism; it breaks up a
signal into different frequencies, just as a prism (or diffraction
grating) breaks up light into different colors (which are light at
different frequencies or wavelengths as c=f).
• A Fourier series is a mathematical depiction of adding overtones to
a basic note to give a richer and fuller sound. It can also be used as
a formula for synthesis of sounds and tones.
• A Fourier series is a representation of x(t) in terms of orthogonal
functions. Sine is orthogonal to cosine. All harmonically related sine
waves are orthogonal to each other. It can be verified by multiplying
the signals over a common period and integrating to get ZERO
result.
Fourier Coefficient Computation
Fourier Series Phasor Representation
Concept of Line Spectrum (Amplitude, Phase Spectrums)
• A line spectrum of a periodic signal x(t) is a visual depiction of its Fourier
coefficients, cn and φn.
• Its amplitude spectrum consists of vertical lines located at discrete values
along the ω-axis, with a line of height c0 located at dc (ω = 0), another of
height c1 at ω = ω0, a third of height c2 at ω = 2ω0, and so on.
• Similarly, the phase spectrum of x(t) consists of lines of lengths
proportional to the values of φn with each located at its corresponding
harmonic nω0.
• Line spectra show at a glance which frequencies in the spectrum of x(t) are
most significant and which are not.
Amplitude and Phase Spectrum
Representation of Periodic Signals in terms of Complex
Exponentials
• A signal is periodic if, for some positive value of T, x(t) = x(t + T) for all t.
• The fundamental period of x(t) is the minimum positive, nonzero value
of T, and the value w = 2/T is referred to as the fundamental
frequency. Higher the frequency, lower the
time period.
=k0 T=T0/k
Consider harmonically related complex exponentials
All have same common period T0.

Assuming that any periodic function may be represented as linear combination of

harmonically related complex exponentials
Since any periodic waveform may be completely
characterized by one period, we can completely
characterize a periodic signal if it may be
represented by linear combination of complex
exponentials having same period as the original
The representation of a periodic signal in the form of equation above signal.
is referred to as the Fourier series representation.
Complex Fourier Series Representation of Periodic Signals
• Equation on above is
referred to as the
synthesis equation.
• Equation for finding
the coefficients is
known as analysis
equation.
• The set of coefficients
ak are often called the
Fourier series
coefficients or the
spectral coefficients
of x(t).
Fourier Series Representation of
Periodic Symmetric Square Wave
Fourier Series of Real Valued Periodic Signal
One Sided and Two-Sided Spectrum
• The line spectra associated with cn and φn are called onesided line
spectra because they are defined for only nonnegative values of
nω0, which follows from the definition of the amplitude/phase
representation, wherein the summation is over only positive values
of n.
• This is to distinguish it from two-sided line spectra associated with
the exponential representation of Fourier Series.
• Exponential representation is commonly used in all signal
processing applications (Particularly in Electronics and
Communications – Two Sided Complex Spectrum) while the
sinusoidal or phasor representation is used in non-intensive signal
processing or inference (One sided Amplitude-Phase Spectrum)
Spectrum
of Real
Valued
Signal
Convergence of Fourier Series
• Approximating a given periodic
signal x(t) by a linear combination of
a finite number of harmonically
related complex exponentials-that is,
by a finite series of the form

The coefficient that minimizes

the error is FS coefficients. The
error tends to zero as N tends to
infinity except at discontinuity
Dirichlet’s Condition
• x(t) equals its Fourier series representation, except at isolated values of t for which x(t)
is discontinuous.
• At these values, the infinite series converges to the average of the values on either side
of the discontinuity.
Convergence and
Gibb’s Phenomenon
Properties of Fourier Series
Properties of Fourier Series
Parseval’s Theorem
Fourier Series Examples
Fourier
Coefficients of
Common
Periodic
Signals
Integration Used in Fourier Series Analysis
Integration Used in Fourier Series Analysis
L-15 and L-16

FOURIER TRANSFORM
From Fourier Series to Fourier Integral and
Fourier Transform
Fourier Series to Fourier Transform
Periodic Pulse Train

The complex Fourier coefficient

multiplied by time period T or Tak is a
continuous function of k0. The
maximum value of this envelope is sin(T1 )
Tak  2T1  2T1sinc(T1 )
the width of the pulse 2T1=.  T1
Time-Frequency Relations of Periodic Pulse Train
• With  thought of as a
continuous variable, the
function (2 sin  T1 )l 
represents the envelope of
Tak
• For fixed T1, the envelope
of Tak is independent of T
• The coefficients ak are
simply equally spaced
samples of this envelope.
The Fourier Transform from Fourier Series
• The Fourier transform may be
regarded as an extension of Fourier
series for a signal with infinite
period but finite energy or
duration. As T  , 0  0 or 0   and k0  k    a continuous variable
• Although Fourier transform is As T  , ak  0 but Tak  X(jk0 )  X(j ) is finite if the integral is finite
developed from Fourier series, it
may be independently thought as a
tool for analysis of any signal or LTI
system in the frequency domain.
– We know that est is an Eigen
function for LTI system and may be
used to obtain homogeneous
solution of any system
represented by a differential
equation.
– Laplace transform gives transform
of any signal or system in terms of
function of s=j as X(s) or H(s).
– The Laplace transform reduces to
Fourier transform as s=j.
Fourier Integral
Forward and Reverse/Inverse Fourier transforms
• The set of Analysis and Synthesis Equations are referred
to as the Fourier transform pair
• X(jw) referred to as the Fourier Transform or Fourier
integral of x(t).
• The synthesis equation is called inverse Fourier transform
equation.
• The synthesis equation plays a role for aperiodic signals
similar to that for periodic signals, since both represent a
signal as a linear combination of complex exponentials.
• For periodic signals, these complex exponentials have
amplitudes ak, which exist at a discrete set of
harmonically related frequencies kw0, k = 0, .± 1, ±2, ....
(Line Spectrum)
• For aperiodic signals, the complex exponentials occur at a
continuum of frequencies and, according to the synthesis
equation (Continuous Spectrum or Spectrum).
Convergence of Fourier Transform
• For the Fourier transform to exist, the integral of the
analysis equation must exist and satisfy Dirichlet’s
condition.
– x(t) must be absolutely integrable as indicated.
– x(t) must have a finite number of maxima and minima within any
finite interval.
– x(t) must have a finite number of discontinuities within any finite
interval. Furthermore, each of these discontinuities must be
finite.
• Fourier transform exist for the following types of
important signals.
– Finite duration and energy signal
– Infinite duration but finite energy or absolutely integrable
signals.
– Complex exponentials or sinusoids – Special treatment is
required and Dirac delta functions must be used as illustrated
later.
Fourier Transform of Decaying Exponential
Magnitude and Phase Spectrum of First Order System
• The first order system impulse
response is of the following type.

The Transform-Pair

If signal represents the impulse

response, the frequency response is

Magnitude and Phase Spectrum

 Fourier Transform of Impulse and Shifted Impulse
x(t )   (t ) X ( j )  2 ( )
  
   1
  ( )e d    ( )e d    ( )d  1
jt  j 0t
X ( j )    (t )e  jt dt    (t )e  j 0 dt    (t )dt  1 x(t ) 
2   
  
For shifted impulses in frequency
For shifted impulses in time
  
X ( j )    (t  t0 )e  jt
dt    (t  t0 )e  jt0
dt  e  jt 0
  (t  t )dt  e
0
 jt 0

  
  
X ( j )    (t  t0 )e  jt dt    (t  t0 )e  j ( t0 ) dt  e jt 0
  (t  t )dt  e
0
jt 0

  

X ( j )  2 (   0 )  e  j 0t
Fourier Transform of Sinusoids
• Using Euler relation and the duality
of shifted impulse, we can establish
a relationship between Fourier
series with Fourier transform (ak
versus X(j)) representations.

Fourier Series and Fourier Transform

Analysis Duality
Rectangular Pulse and Sinc( ) Function


sin(Wt )  jt
X ( j )   t e dt
Duality of Rectangular Function and Sinc Function
Fourier Transform Pairs
(Tables of FT Pairs)
• For many cases, we use the
duality properties or Fourier
transform pairs in obtaining
Fourier transform or inverse
Fourier transform.
Properties of Fourier Transform
Linearity Conjugation and Conjugate Symmetry

Time-Shifting

if x(t) is real, then X(jw) has conjugate

symmetry
Frequency-Shifting

The effect of a time shift on a signal is to introduce into its transform

a phase shift, namely, -wt0 , which is a linear function of w.
Properties of Fourier Transform
Differentiation

Integration
Convolution and Multiplication in Frequency Domain

Convolution is multiplication in frequency

domain. This relation has a major importance
in signal and system analysis

Multiplication in time domain or modulation

by duality is convolution in the frequency
domain.
Modulation or Multiplication in Time Domain
Properties of Fourier Transform
Time and Frequency Scaling

substitution  = at,

The shorter the pulse, the broader is the

spectrum or vice-versa.
Parseval’s Identity and Energy Density Spectrum
• The term on the left-hand side) is the total
energy in the signal x(t).
• Parseval's relation says that
– Total energy of the signal may be determined by
integrating energy per unit frequency in the
frequency domain.
• For this reason, |X(jw )|2 is often referred to
as the energy-density spectrum of the signal
x(t).
• Parseval's relation for finite-energy signals is
the direct counterpart of Parseval's relation
for periodic signals, which states that the
average power of a periodic signal equals the
sum of the average powers of its individual
harmonic components, which in turm are
equal to the squared magnitudes of the
Fourier series coefficients.
– The power spectral density of periodic signal
LTI System Analysis Using Fourier Method
• Steady state response of LTI system may be
analyzed using Fourier series
• Obtain Fourier series representation of the
periodic signal.
• Obtain frequency response of the system by
Fourier transforming the impulse response.
• Obtain the response for each sinusoid or
complex exponential component.
• Sum or use superposition to obtain the
response for the periodic signal.
• The response y(t) is also periodic with the
same fundamental frequency as x(t).
• If ak is the set of Fourier series coefficients for
the input x(t), then {akH(jkwo)} is the set of
Fourier coefficients for the output y(t).
Example
Frequency Response from Differential Equation
Using Eigen function, particular solution may be obtained
Using Fourier Transform [First Order]
Using differentiation
property of FT

For impulse input x(t)=(t) or X(jw) = 1

2nd Order System Response
Supplements
THEORY FOR FOURIER SERIES AND FOURIER
TRANSFORM DEVELOPMENT
COMPLEX EXPONENTIAL AS EIGEN FUNCTION
TRANSFER FUNCTION OR FREQUENCY RESPONSES AS EIGEN VALUES
Transform Domain Analysis Principle Complex System Scaled
Fourier, Laplace, and Z-Transform Exponential
Input
(Transfer
Function H)
Complex
Exponential
Output
• The response of an LTI system to a
complex exponential input is the
same complex exponential with
only a change in amplitude.
– The complex amplitude factor H(s) or
H(z) will in general be a function of
the complex variable s or z.
• A signal for which the system
output is a (possibly complex)
constant times the input is referred
to as an Eigen Function of the
system, and the amplitude factor is
referred to as the system's Eigen
Value.
Complex Exponentials as Eigen Functions of CT LTI Systems
• Analysis of system output for complex
exponential input x(t)=est where s=+j is
a complex number.
• The analysis shows that x(t)=est as input
produces output of type est which is scaled
by the factor H(s).
• Hence, est is an eigen function of an LTI
system and H(s) is the Eigen Value.
• Linear combination of Eigen function
inputs will different parameters s produce
linear combination of outputs as per the
superposition property.
• Hence, it will be easier to find response of
a input as linear combination of responses
to individual components.
Complex Exponentials as Eigen Functions of DT LTI Systems
• Analysis of system output for complex
exponential input x[n]=zn where z is a
complex number.
• The analysis shows that x[n]=zn as input
produces output of type zn which is
scaled by the factor H(z).
• Hence, zn is an eigen function of an LTI
system and H(z) is the Eigen Value.
• Linear combination of Eigen function
inputs will different parameters z
produce linear combination of outputs as
per the superposition property.
• Hence, it will be easier to find response
of a input as linear combination of
responses to individual components.