DIGITAL SIGNAL PROCESSING

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Lecture 13: Summary
• Properties of Commonly used Windows
• Kaiser Window Filter Design Method
• The Discrete Fourier Series
• The Discrete Fourier Series Example
• Properties of Discrete Fourier Series

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Lecture 14: Outline
• The Fourier Transform of Periodic Signals
• Sampling the Fourier Transform
• The Discrete Fourier Transform
• Properties of DFT
• Circular Convolution
• Linear Convolution vs Circular Convolution

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 The Fourier Transform of
Periodic Signals
• Periodic sequences are not absolute or square
summable
• Hence they don’t have a Fourier Transform
• We can represent them as sums of complex
exponentials: DFS
• We can combine DFS and Fourier transform
• Fourier transform of periodic sequences
• Periodic impulse train with values proportional to
DFS coefficients ~ j 2 ~ 
Xe   

2k 
Xk    
k   N  N 
• This is periodic with 2 since DFS is periodic
Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing
 The Fourier Transform of
Periodic Signals
• The inverse transform can be written as
1 2    ~ j jn 1 2    2 ~ 
2   
X e e d 
2   Xk   
2k  jn
e d
k   N  N 
0   0  

2 k
1  ~ 2    2k  jn 1 N1 ~
Xk  e d   Xk e
j n

N k   0  
  
 N  N k 0
N

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 The Fourier Transform of
Periodic Signals
EXAMPLE
• Consider the periodic impulse train

~
p[n]   n  rN
r  

• The DFS was calculated previously to be

~
P k   1 for all k
• Therefore the Fourier transform is
~ j
  2  2k 

Pe  
k   N
  
 N 


Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 The Fourier Transform of
Periodic Signals
Relation Between Finite Length and Periodic Sequences
• Consider finite length signal x[n] spanning from 0 to
N-1
• Convolve with periodic impulse train

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 The Fourier Transform of
Periodic Signals
Relation Between Finite Length and Periodic Sequences
• Consider finite length signal x[n] spanning from 0 to
N-1
• Convolve with periodic impulse train
 
x[n]  x[n]  ~
~ p[n]  x[n]   n  rN   xn  rN
r   r  

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 The Fourier Transform of
Periodic Signals
Relation Between Finite Length and Periodic Sequences
• The Fourier transform of the periodic sequence is
~ j ~
        2N    2Nk 

X e  X e j P e j  X e j
k    
2  j N  
2 k
~ j
  2k 

Xe  
k   N
X e   
N 

  

• This implies that

~  j 2Nk 
Xk   X e   
  X e j

2 k
  N

• DFS coefficients of a periodic signal can be thought

as equally spaced samples of the Fourier transform of
one period
Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing
 The Fourier Transform of
Periodic Signals
EXAMPLE
• Consider the following
sequence
1 0  n  4
x[n]  
0 else

• The Fourier transform

sin5 / 2
Xe j   e j2
sin / 2

• The DFS coefficients

~ sink / 2
Xk   e j4 k / 10 
sink / 10
Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing
 Sampling the Fourier
Transform
• Consider an aperiodic sequence with a Fourier
transform
x[n] DTFT
 X e j  
• Assume that a sequence is obtained by sampling the
~
DTFT
Xk   X e j  
 X e j2 / Nk
  2  / Nk
 
• Since the DTFT is periodic resulting sequence is also
periodic
• We can also write it in terms of the z-transform
(
X [ k ] = X ( z) z=e j(2 p /N )k = X e
j ( 2 p /N ) k
)
Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing
 Sampling the Fourier
Transform
• The sampling points are shown in figure
• could be the DFS of a sequence
• Write the corresponding sequence
~ 1 N1 ~
x[n]   Xk e j2  / Nkn
N k 0

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Sampling the Fourier
Transform
• The only assumption made on the sequence is that
DTFT exist
Xe    xme

~
Xk   Xe  x[n]   ~Xk e j2 / Nkn
j  jm j2  / Nk ~ 1 N 1

m   N k 0

• Combine equation to get

~ 1 N1    j2  / N km  j2  / N kn

x[n]     xme  e
N k  0 m   

 1 N1 j2  / Nk nm  
~
  xm  e    xmp n  m
m   N k  0  m  

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Sampling the Fourier
Transform
• Term in the parenthesis is
~ 1 N1 j2  / Nk n m 
p n  m   e   n  m  rN
N k 0 r  

• So we get
 
~
x[n]  xn   n  rN   xn  rN
r   r  

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Sampling the Fourier
Transform

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Sampling the Fourier
Transform
• Samples of the DTFT of an aperiodic sequence can be
thought of as
• DFS coefficients of a periodic sequence obtained
through summing periodic replicas of original
sequence
• If the original sequence is of finite length
• and we take sufficient number of samples of its
DTFT
• the original sequence can be recovered by
~x n 0  n  N  1
xn  
 0 else

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Sampling the Fourier
Transform
• It is not necessary to know the DTFT at all
frequencies
• To recover the discrete-time sequence in time
domain
• Discrete Fourier Transform
• Representing a finite length sequence by samples
of DTFT

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 The Discrete Fourier
Transform
• Consider a finite length sequence x[n] of length N
xn  0 outside of 0  n  N  1
• For given length-N sequence associate a periodic

sequence ~
x n  xn  rN 
r  

• The DFS coefficients of the periodic sequence are

samples of the DTFT of x[n]
• Since x[n] is of length N there is no overlap between
terms of x[n-rN] and we can write the periodic
sequence as
~
xn  xn mod N  xnN 
Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing
 The Discrete Fourier
Transform
• To maintain duality between time and frequency
~
• We choose one period of Xk  as the Fourier
transform of x[n]
~
Xk  0  k  N  1

Xk    ~
Xk   Xk mod N  Xk N 

 0 else

• The DFS pair

~ 1 N1 ~
~ x[n]   Xk e j2  / Nkn
N 1
Xk   ~
 x
n0
[n]e  j2  / N kn
N k 0

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 The Discrete Fourier
Transform
• The equations involve only on period so we can write
N1 ~
 x[n]e  j2  / Nkn 0  k  N  1
Xk   n 0
 0 else
1 ~ N 1
  Xk e j2  / Nkn 0  k  N  1
x[n]  N k  0
 0 else
• The Discrete Fourier Transform
N 1
1 N1
Xk    x[n]e  j2  / N kn
x[n]   Xk e j2  / Nkn
n0 N k 0
• The DFT pair can also be written as
x[n]¬¾¾
DFT
® X [k]
Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing
 The Discrete Fourier
Transform
EXAMPLE
•The DFT of a rectangular pulse
•x[n] is of length 5
•We can consider x[n] of any
length greater than 5
•Let’s pick N=5
•Calculate the DFS of the periodic
form of x[n]
4
1- e- j 2 p k
X [k ] = åe
- j ( 2 p k/5) n
=
1- e (
- j 2 p k/5)
n=0

ìï 5 k = 0,±5,±10,...
=í
ïî 0 else
Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing
 The Discrete Fourier
Transform
EXAMPLE
• If we consider x[n] of length 10
• We get a different set of DFT
coefficients
• Still samples of the DTFT but in
different places

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Properties of the Discrete
Fourier Transform
• Linearity
x1 n DFT
 X1 k 
x2 n DFT
 X2 k 
ax1 n  bx2 n DFT
 aX1 k   bX2 k 

• Circular Shift of Sequence

xn DFT
 Xk 
xn  mN  0  n  N - 1 DFT
 Xk e  j2 k / Nm

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Properties of the Discrete
Fourier Transform
• Duality xn DFT
 Xk 
Xn DFT
 Nx k N 

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Properties of the Discrete
Fourier Transform
• Symmetry Properties

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Properties of the Discrete
Fourier Transform
Circular Convolution
• Circular convolution of
of two finite length
sequences
N 1
x3 n   x mx n  m 
1 2 N
m0

N 1
x3 n   x mx n  m 
2 1 N
m0

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Properties of the Discrete
Fourier Transform
EXAMPLE: Circular
Convolution 1 0  n  L  1
x1 n  x2 n  
• Circular convolution of two 0 else
rectangular pulses L=N=6 N 1 2
N k  0
X1 k   X2 k  
j kn
e N

n0 0 else
• DFT of each sequence
N2 k  0
X3 k   X1 k X2 k   
 0 else
• Multiplication of DFTs
N 0  n  N  1
x3 n  
0 else
• And the inverse DFT
Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing
 Properties of the Discrete
Fourier Transform
EXAMPLE: Circular
Convolution
• Circular convolution of two
rectangular pulses L=N=6

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Properties of the Discrete
Fourier Transform
EXAMPLE: Circular Convolution
• We can augment zeros to each sequence L=2N=12
• The DFT of each sequence
2 Lk
j
1e N
X1 k   X2 k   2 k
j
1e N

• Multiplication of DFTs
2
 j
2 Lk

1  e N 
X3 k    2 k 
1e N j

 

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Properties of the Discrete
Fourier Transform
EXAMPLE: Circular
Convolution

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Circular Convolution
Illustration of Circular Convolution Process

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Circular Convolution
Example

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Linear Vs Circular
Convolution
Linear Convolution of Two Sequence of Length L and P

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Linear Vs Circular
Convolution
Linear Convolution of Two Equal Length Sequence
L=P=6

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Linear Vs Circular
Convolution
Linear Convolution of Two Equal Length Sequence
L=P=6

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Linear Vs Circular
Convolution
Circular Convolution of Two Equal Length Sequence
with N=L=6, and N=2L=12

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Linear Vs Circular
Convolution
Linear Convolution of Two Different Length Sequence L
and P

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Linear Vs Circular
Convolution
Circular Convolution of Two Different Length Sequence
L and P

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Linear Vs Circular
Convolution
Circular Convolution of Two Different Length Sequence
L and P

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing

 Linear Vs Circular
Convolution
Circular Convolution of Two Different Length Sequence
L and P

Muhammad Majid (m.majid@uettaxila.edu.pk) Digital Signal Processing