TON DUC THANG UNIVERSITY

FACULTY OF ELECTRICAL AND

ELECTRONICS ENGINEERING

402070
DIGITAL SIGNAL PROCESSING
CHAPTER 7: THE DISCRETE FOURIER TRANSFORM

Tan N. Nguyen, Msc.

 OBJECTIVES

 Understand the discrete Fourier transform.

 Know the properties of the discrete Fourier
transform
 Understand the discrete cosine transform

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 2

 CHAPTER 7: THE DISCRETE
FOURIER TRANSFORM

 7.1 Representation of Periodic Sequence: the

Discrete Fourier Series
 7.2 Properties of the Discrete Fourier Series
 7.3 The Fourier Transform of Periodic Signal
 7.4 Fourier Representation of Finite-Duration
Sequence: the Discrete Fourier Transform
 7.5 Properties of the Discrete Fourier Transform
 7.6 The Discrete Cosine Transform

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 3

 THE DISCRETE FOURIER SERIES

~
x[n] with
 Given a periodic sequence period N
so that ~
x[n]  ~x[n  rN]
 The Fourier series representation can be
written as 1 j  2 / N kn
x[n]   X k  e
N k
 Fourier series representation of continuous-
time periodic signals require infinite many
complex exponentials

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 4

 THE DISCRETE FOURIER SERIES
 Not that for discrete-time periodic signals we
have
j2  / N k  mN n j2  / N kn j2 mn  j2  / N kn
e e e e
 Due to the periodicity of the complex
exponential we only need N exponentials for
discrete time Fourier series
1 N 1
x[n]   X  k  e j  2 / N kn
N k 0
1 j  2 / N kn
 No need x[n]   X  k  e
N k
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 5
 THE DISCRETE FOURIER SERIES
 Discrete Fourier Series Pair
 A periodic sequence in terms of Fourier series
coefficients 1 N 1 j  2 / N  kn
x[n]   X  k  e
N k 0
 To obtain the Fourier series coefficients we
multiply both sides by e j (2 / N ) rn for 0nN-1
and then sum both the sides , we obtain
N 1 2 N 1 N 1 2
j 1 ( k r ) n
    X (k )e
rn j
x(n)e N N
n 0 n 0 N k 0
N 1 2 N 1 N 1 2
j 1 ( k r ) n
   X ( k ) e
rn j
x(n)e N N
n 0 k 0 n 0 N
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 6
 THE DISCRETE FOURIER SERIES
1 N 1
j  2 / N  kn
x[n] 
N
 X k  e
k 0
2
N 1 j N 1
1 j 2N ( k r ) n
N 1

 x(n)e   X ( k ) e
rn
N
n 0 k 0 n 0 N
2
1 N 1 ( k r ) n 1, k - r  mN , m an integer
e
j
N 
N n 0 0, otherwise
N 1 2
j
 x(n)e
rn
N  X (r )
n 0
N 1 2
j
X (k )   x(n)e
kn
N
n 0
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 7
 THE DISCRETE FOURIER SERIES

 Representation of Periodic Sequence: the

Discrete Fourier Series
 x n with period N,
a periodic sequence ~
x n  ~
~ x n  rN  for any integer r
 The Fourier series coefficients of ~x n is
N 1
 j  2 N kn
X  k    x  n e
n 0

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

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 8

 THE DISCRETE FOURIER SERIES
N 1
 j  2 N kn
X  k    x  n e
n 0

The sequence X k  is periodic with period N

~

X 0  X N , X 1  X N  1
~ ~ ~ ~
N 1
 j  2 N  k  N n
X  k  N    x n e
n 0

 N 1  j  2 N kn   j 2 n
   x  n e e  X k 
 n 0 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 9
 THE DISCRETE FOURIER SERIES

 Discrete Fourier Series (DFS)

N 1
 j  2 N kn  j  2 N 
X  k    x  n e  Let WN  e
n 0
N 1
X k    x nWN
~ ~ kn
Analysis equation:
n 0
N 1
x n   X k WN
~ 1 ~  kn
Synthesis equation:
DFS N k 0
x n  X k 
~ ~ F
discrete  periodic
F
periodic  discrete
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 10
 THE DISCRETE FOURIER SERIES

 Example: DFS of a impulse train

 Consider the periodic impulse train

1, n  rN , r is any integer
x n    n  rN   
~
r   0, therwise
x n
N points ~

-N -N+1…… -2 -1 0 1 2 …… N-1 N N+1 N+2 ……

N 1
X k     nWNkn  WN0  1
~
n 0
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 11
 THE DISCRETE FOURIER SERIES
N 1
X k     nWN  WN  1
~ kn 0

n 0
X k 
N points ~

k
-N -N+1…… -2 -1 0 1 2 …… N-1 N N+1 N+2 ……

 1, n  rN , r is any integer
x  n      n  rN   
r  0, therwise
1 N 1 1 N 1
  X  k WN   e
 kn j  2 N  kn

N k 0 N k 0
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 12
 THE DISCRETE FOURIER SERIES
N points
1 x n
~

-N -N+1…… -2 -1 0 1 2 …… N-1 N N+1 N+2 ……k

X k 
N points ~
1

-N -N+1…… -2 -1 0 1 2 …… N-1 N N+1 N+2 …… n

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 13

 THE DISCRETE FOURIER SERIES

 Example: The Discrete Fourier Series of a

Periodic Rectangular Pulse Train
 Periodic sequence with period N=10

1W
4 4
 j  2 10  kn
X  k   W  e
5k
kn
10  10
n 0 n 0 1W k
10
 4 k 10 sin  k 2 
e
sin  k 10 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 14
 THE DISCRETE FOURIER SERIES

magnitude

 4 k 10 sin  k 2 
phase X k   e
sin  k 10 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 15

 THE DISCRETE FOURIER SERIES

magnitude
 4 k 10 sin  k 2 
X k   e
sin  k 10 

phase

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 16

 CHAPTER 7: THE DISCRETE
FOURIER TRANSFORM

 7.1 Representation of Periodic Sequence: the

Discrete Fourier Series
 7.2 Properties of the Discrete Fourier Series
 7.3 The Fourier Transform of Periodic Signal
 7.4 Fourier Representation of Finite-Duration
Sequence: the Discrete Fourier Transform
 7.5 Properties of the Discrete Fourier Transform
 7.6 The Discrete Cosine Transform

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 17

 PROPERTIES OF THE DISCRETE
FOURIER SERIES
 Linearity: two periodic sequence, both with
period N DFS
x1 n  X 1 k 
~ ~

DFS
x2 n  X 2 k 
~ ~

DFS
ax1 n  bx2 n  aX 1 k   bX 2 k 
~ ~ ~ ~

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 18

 PROPERTIES OF THE DISCRETE
FOURIER SERIES
 Shift of a sequence
DFS
x n  X k 
~ ~

DFS
x n  m  WN X k 
~ km ~

DFS
WN nl x  n  X  k  l 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 19

 PROPERTIES OF THE DISCRETE
FOURIER SERIES
 Duality
DFS
x n  X k 
~ DFS
X n  N~x  k 
~ ~

x n
~
X k 
~
1
1

n
0 1 2 …… N-1 …… N-1
k
0 1 2

X  n
1 N
Nx  k 
k
0 1 2 …… N-1
n
0 1 2 …… N-1

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 20

 CHAPTER 7: THE DISCRETE
FOURIER TRANSFORM

 7.1 Representation of Periodic Sequence: the

Discrete Fourier Series
 7.2 Properties of the Discrete Fourier Series
 7.3 The Fourier Transform of Periodic Signal
 7.4 Fourier Representation of Finite-Duration
Sequence: the Discrete Fourier Transform
 7.5 Properties of the Discrete Fourier Transform
 7.6 The Discrete Cosine Transform

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 21

 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL
 Periodic sequences are neither absolutely

summable nor square summable, hence they

don’t have a strict Fourier Transform
    2  w  2 r 

xn  1 for all n 
F jw
X e
r  

xn  e X e jw  


 2  w  w  2 r 
jw0 n F
 0
r  

X e jw  

x  n   ak e  2 a  w  w  2 r 
jwk n F
 k k
k r   k

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 22

 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL
 We can represent Periodic sequences 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
1 N 1
j  2 N  kn
x  n   X k  e
N k 0
2 
2 k 

X e   
j
X  k    - 
k - N  N 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 23
 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL
2

 2 k 
X e   
j
X  k    - 
k - N  N 
This is periodic with 2 since DFS is periodic
 The inverse transform can be written as

2  2 k  jn
X  e e d 
1 2 - 1 2 -

j j n
  X  k    - e d
2 0- 2 0-
k - N  N 
2 k
1 
 2 k  jn
2 - 1 N -1
  X  k     - e d   X  k  e
j n

N
N k -  0-
 N  N k 0
 x  n

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 24

 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL
 Example: Fourier Transform of a periodic
impulse train
 Consider

the periodic impulse train
p[n]     n  rN 
r 
1
N points
p  n
… … …
-N … -2 -1 0 1 2… N …
The DFS was calculated previously to be
N points P k 
1
P  k   1 for all k … … …
-N … -2 -1 0 1 2 … N-1 N … n
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 25
 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL

Therefore the Fourier transform is

2 

2 k 
P e   
j
   
k  N  N 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 26

 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL
 Relation between Finite-length and Periodic
Signals
 Consider finite length signal x[n] spanning from 0
to N-1
p  n
1
… …
-N -2 -1 0 1 2 N
… …
 Convolve with periodic impulse train

 
x[n]  x[n]  p[n]  x[n]     n  rN    x n  rN 
r  r 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 27
 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL
 The Fourier transform of the periodic
sequence is
2  2 k  
X e   X e  P e   X e  
j j j
   j

k  N  N 
2 k

2    2 k 
X e   
j
   
j
X e N

k  N    N 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 28

 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL

2  2 k 
X e   
j
X  k    - 
k - N  N 
2 k

2    2 k 
X e   
j
   
j
X e N

k  N    N 
 j 2N k 
This implies that X  k   X  e   X     2 k
e j

  N

DFS coefficients of a periodic signal can be

thought as equally spaced samples of the Fourier
transform of one period

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 29

 THE FOURIER TRANSFORM OF
PERIODIC SIGNAL
 x n is periodic with period N, the DFS are
If ~
N 1
X k    ~ x ne  j 2
~ N kn

n 0

 If xn  ~x n for 0  n  N 1 and xn  0 otherwise

then
 
N 1 N 1
X e   xne
jw  jwn
  x ne
~  jwn

n 0 n 0

X k   X e jw
 w2 k N
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 30
 CHAPTER 7: THE DISCRETE
FOURIER TRANSFORM

 7.1 Representation of Periodic Sequence: the

Discrete Fourier Series
 7.2 Properties of the Discrete Fourier Series
 7.3 The Fourier Transform of Periodic Signal
 7.4 Fourier Representation of Finite-Duration
Sequence: the Discrete Fourier Transform
 7.5 Properties of the Discrete Fourier Transform
 7.6 The Discrete Cosine Transform

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 31

 THE DISCRETE FOURIER
TRANSFORM
 Consider a finite-length sequence xn of
length N samples such that xn  0 outside
the range 0  n  N 1
N 9

 To each finite-length sequence of length N,

we can associate a period sequence

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 32

 THE DISCRETE FOURIER
TRANSFORM

x  n   x  n  rN 
r 
 x n, 0  n  N  1
~
xn  
 0, otherwise

x  n  x  n mod N   x   n   N 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 33

 THE DISCRETE FOURIER
TRANSFORM
For x n, the DFS is X k  with period N
~ ~


 The Discrete Fourier Transform of xn is

 X k , 0  k  N  1
~
X k   
 0, otherwise

X  k   X  k mod N    X   k   N 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 34

 THE DISCRETE FOURIER
TRANSFORM
N 1 N 1
x n   
1
X k    x nWN
~
~ ~ kn ~
N
 X
k 0
k W  kn
N
n 0

N 1
 xnWNkn , 0  k  N  1
X k    n0

 0, otherwise

 1 N 1
  X k WNkn , 0  n  N  1
xn   N k 0

 0, otherwise

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 35

 THE DISCRETE FOURIER
TRANSFORM
 Analysis equation
N 1
X k    xnW kn
N
n 0

 Synthesis equation
N 1
xn   X k W
1  kn
N
N k 0

DFT
xn  X k 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 36

 THE DISCRETE FOURIER
TRANSFORM
Time Frequency
Fourier transform continuous continuous
(FT)
Fourier series (FS) continuous Continuous
periodic impulse train
Discrete-time discrete continuous
Fourier transform periodic
(DTFT)
Discrete Fourier discrete continuous impulse
series (DFS) periodic train, periodic
Discrete Fourier discrete discrete
transform (DFT)
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 37
 CHAPTER 7: THE DISCRETE
FOURIER TRANSFORM

 7.1 Representation of Periodic Sequence: the

Discrete Fourier Series
 7.2 Properties of the Discrete Fourier Series
 7.3 The Fourier Transform of Periodic Signal
 7.4 Fourier Representation of Finite-Duration
Sequence: the Discrete Fourier Transform
 7.5 Properties of the Discrete Fourier Transform
 7.6 The Discrete Cosine Transform

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 38

 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
 Linearity
DFT
ax1 n  bx2 n  aX 1 k   bX 2 k 

If x1n has length N1 and x2 nhas length N 2,

N3  max N1 , N 2 
N3 1
X1  k    1   N3 , 0  k  N 3  1
x n W kn

n 0
N3 1
X 2 k    2   N3 , 0  k  N 3  1
x n W kn

n 0
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 39
 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
 Circular Shift of a Sequence
DFS
x n  X k 
~ ~

DFS
 j  2 k N m
x  n  m  e X k 

DFT
xn, 0  n  N  1  X k 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 40

 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM

DFT
 j  2 k N m
x   n  m   N  , 0  n  N  1  e X k 

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 41

 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
xn  X e  jw x  n  m  e  jwm
X  e jw 
DFS DFS
 j  2 k N m
x n  X k  x  n  m  e X k 
~ ~

DFT
xn, 0  n  N  1  X k 
DFT
 j  2 k N m
x1  n , 0  n  N  1  X1  k   e X k 

x  n  x   n   N   X  k   X   k   N 
DFS

x1  n  x1   n   N   X1  k   X1   k   N 
DFS

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 42

 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
x  n  x   n   N   X  k   X   k   N 
DFS

x1  n  x1   n   N   X1  k   X1   k   N 
DFS

 j  2 k N m
X1 k   e X k 
 
 j  2   k   N m
X1  k   e X   k   N 
  N 

 j  2 k N m  j  2 k N m
e X   k   N   e X k 
 j  2 k N m
x1  n  x   n  m   N   x  n  m
DFS

e X k 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 43
 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
 j  2 k N m
X1 k   e X k 
 j  2 k N m
x1  n  x   n  m   N   x  n  m  e
DFS
  X k 

 x1  n  x   n  m   N  , 0  n  N  1

x1  n  

 0, otherwise
DFT
xn, 0  n  N  1  X k 
DFT
xn  mN , 0  n  N  1  e  j 2 k N m
X k 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 44
 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
DFS
x n  X k 
~ ~
 Duality
x  n  x   n   N   X  k   X   k   N 
DFS

x1  n  X  n X1  k   Nx  k 
DFS
X n  N~x  k 
~

x1  n  X  n
 Nx  k   Nx   k   N  , 0  k  N  1

X1  k   

 0, otherwise
DFT
xn X k  Nx   k   N  , 0  k  N  1
DFT

 X  n 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 45
 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
 Symmetry Properties
x  n  x   n   N   X  k   X   k   N 
DFS

DFT
x* n  X *  k N , 0  n  N  1
DFT
x*  n N   X * k , 0  n  N  1

xe  n    x  n   x*  n 
1
2
xo  n    x  n   x  n 
1 *

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 46

 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
e
2
 N   
xep  n  x  n   1 x   n     x*   n    ,0  n  N 1
  N

xop n  x  n    x   n     x   n   ,0  n  N 1

1 *
o N N
2
for 0  n  N  1,  n  N  N  n,  n N  n
xep  n    x  n   x*  N  n  , 0  n  N  1
1
2
xep  0   x  0  x  N    x  0  x*  0  Re  x  0
1 * 1
2 2
1
 
xop  n   x  n   x*  N  n  , 0  n  N  1
2
x0 p  0   x  0  x*  0  j Im  x  0
1
2
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 47
 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
xep  n    x  n   x  N  n  , 0  n  N  1
1 *

2
xop  n    x  n   x*  N  n  , 0  n  N  1
1
2
xe  n    x  n   x  n  xo  n    x  n   x  n 
1 *
1 *

2 2
xep  n    xe  n   xe  n  N  , 0  n  N  1

xop  n    xo  n   xo  n  N  , 0  n  N  1

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 48

 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
x  n  xe  n  xo  n
x  n  x n  xe n  xo n , 0  n  N  1
xep  n  xe  n , 0  n  N 1
xop  n  xo  n , 0  n  N 1
x  n  xep  n  xop  n
DFT
j Imxn  X op k 
DFT
Rexn  X ep k 
DFT
xop n  j ImX k 
DFT
xep n  ReX k 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 49
 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 50

 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
 Circular Convolution
N 1
x 
1 2 m  x  n  m  DFS
 X1  k  X 2  k 
m0

For two finite-duration sequences x1  n and x2  n,

both of length N, with DFTs X1  k  and X 2  k 
X 3  k   X1  k  X 2  k 
N 1
x3  n    x1  m  x2  n  m , 0  k  N  1
m0
N 1
x3  n    x1   m   N  x2   n  m   N , 0  k  N  1
m0
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 51
 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
N 1
x3  n    x1   m   N  x2   n  m   N , 0  k  N  1
m0
N 1
  x1  m  x2   n  m   N , 0  k  N  1
m0

since  m N
 m, for 0  k  N  1
N 1
x3  n    x1  m  x2   n  m   N 
m0

 x1  n  N x2  n 
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 52
 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
N 1
x3  n    x1  m  x2   n  m   N 
m0

 x1  n  N x2  n   x2  n N x1  n
N 1
  x2  m x1   n  m   N 
m 0

DFT
x3 n  X 3 k   X 1 k X 2 k , 0  k  N  1

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 53

 PROPERTIES OF THE DISCRETE
FOURIER TRANSFORM
DFT
x1 n N x2 n  X 1 k X 2 k 
if x3 n  x1 nx2 n
N 1
X 3 k    X l X k  l  
1
1 2 N
N l 0

DFT
x1 nx2 n  X 1 k  N X 2 k 
1
N

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 54

 CHAPTER 7: THE DISCRETE
FOURIER TRANSFORM

 7.1 Representation of Periodic Sequence: the

Discrete Fourier Series
 7.2 Properties of the Discrete Fourier Series
 7.3 The Fourier Transform of Periodic Signal
 7.4 Fourier Representation of Finite-Duration
Sequence: the Discrete Fourier Transform
 7.5 Properties of the Discrete Fourier Transform
 7.6 The Discrete Cosine Transform

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 55

 THE DISCRETE COSINE
TRANSFORM
 Definition DCT-1

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 56

 THE DISCRETE COSINE
TRANSFORM
 DCT-2

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 57

 THE DISCRETE COSINE
TRANSFORM
 Relationship between DCT-1 and DFT

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 58

 THE DISCRETE COSINE
TRANSFORM
 Relationship between DCT-2 and DFT

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 59

 THE DISCRETE COSINE
TRANSFORM
 Energy Compaction Property of the DCT-2

x(n)  a cos(0 n   )
n

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 60

 SUMMARY

 This chapter we have learnt :

 Discrete Fourier Transform (DFT) for finite
duration sequence
 DFT is a sequence rather than a function of a
continuous variable
 DFT corresponds to sample, equally spaced in
frequency, of the Fourier transform of the signal.
 The Fourier series representation of the periodic
sequence corresponds to the DFT of the finite-
length sequence.
23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 61
 HOMEWORK

 Review chapter 7
 Read chapter 8(pp. 716->742)
 Solve the exercises : 8.3, 8.4, 8.7, 8.10, 8.51,
8.52, 8.53.

23/05/2016 402070-Chapter 7: The Discrete Fourier Transform 62