The Discrete Fourier Transform

Content and Figures are from Discrete-Time Signal Processing, 2e by Oppenheim, Shafer, and Buck, 1999-2000 Prentice Hall
Inc.
Spring 2008 2
Sampling the Fourier Transform
Consider an aperiodic sequence with a Fourier transform

Assume that a sequence is obtained by sampling the DTFT


Since the DTFT is periodic resulting sequence is also periodic
We can also write it in terms of the z-transform


The sampling points are shown in figure
could be the DFS of a sequence
Write the corresponding sequence



| | ( )
( )
( )
( )
k N / 2 j
k N / 2
j
e X e X k X
~
t
t = e
e
= =
( )
e

j DTFT
e X ] n [ x
| | ( )
( )
( )
( )
k N / 2 j
e z
e X z X k X
~
k N / 2
t
=
= =
t
| | k X
~
| |
( )

=
t
=
1 N
0 k
kn N / 2 j
e k X
~
N
1
] n [ x
~
Spring 2008 3
Sampling the Fourier Transform Contd
The only assumption made on the sequence is that DTFT exist


Combine equation to get





Term in the parenthesis is


So we get

( ) | |

=
e e
=
m
m j j
e m x e X
| |
( )

=
t
=
1 N
0 k
kn N / 2 j
e k X
~
N
1
] n [ x
~
| |
( )
( )
k N / 2 j
e X k X
~
t
=
| |
( ) ( )
| |
( ) ( )
| | | |

=
t

=
t

=
t
=
(

=
(

=
m m
1 N
0 k
m n k N / 2 j
1 N
0 k
kn N / 2 j
m
km N / 2 j
m n p
~
m x e
N
1
m x
e e m x
N
1
] n [ x
~
| |
( ) ( )
| |


=

=
t
o = =
r
1 N
0 k
m n k N / 2 j
rN m n e
N
1
m n p
~
| | | | | |


=

=
= o - =
r r
rN n x rN n n x ] n [ x
~
Spring 2008 4
Sampling the Fourier Transform Contd

Spring 2008 5
Sampling the Fourier Transform Contd
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




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
| |
| |

s s
=
else 0
1 N n 0 n x
~
n x
Spring 2008 6
The Discrete Fourier Transform
Consider a finite length sequence x[n] of length N

For given length-N sequence associate a periodic sequence


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


To maintain duality between time and frequency
We choose one period of as the Fourier transform of x[n]


| | 1 N n 0 of outside 0 n x s s =
| | | |

=
=
r
rN n x n x
~
| | ( ) | | ( ) ( ) | |
N
k X N mod k X k X
~
= =
| | k X
~
| |
| |

s s
=
else 0
1 N k 0 k X
~
k X
| | ( ) | | ( ) ( ) | |
N
n x N mod n x n x
~
= =
Spring 2008 7
The Discrete Fourier Transform Contd
The DFS pair


The equations involve only on period so we can write






The Discrete Fourier Transform


The DFT pair can also be written as
| |
( )

=
t
=
1 N
0 k
kn N / 2 j
e k X
~
N
1
] n [ x
~
| |
( )

=
t
=
1 N
0 n
kn N / 2 j
e ] n [ x
~
k X
~
| |
( )

s s
=

=
t
else 0
1 N k 0 e ] n [ x
~
k X
1 N
0 n
kn N / 2 j
| |
( )

s s
=

=
t
else 0
1 N k 0 e k X
~
N
1
] n [ x
1 N
0 k
kn N / 2 j
| |
( )

=
t
=
1 N
0 n
kn N / 2 j
e ] n [ x k X
| |
( )

=
t
=
1 N
0 k
kn N / 2 j
e k X
N
1
] n [ x
| | ] n [ x k X
DFT

Spring 2008 8
Example
The DFT of a rectangular pulse
x[n] is of length 5
We can consider x[n] of any
length greater than 5
Lets pick N=5
Calculate the DFS of the
periodic form of x[n]
| |
( )
( )

=
=

=
=
t
t
=
t

else 0
,... 10 , 5 , 0 k 5

e 1
e 1

e k X
~
5 / k 2 j
k 2 j
4
0 n
n 5 / k 2 j
Spring 2008 9
Example Contd
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

Spring 2008 10
Properties of DFT
Linearity




Duality





Circular Shift of a Sequence


| | | |
| | | |
| | | | | | | | k bX k aX n bx n ax
k X n x
k X n x
2 1
DFT
2 1
2
DFT
2
1
DFT
1
+ +


| | | |
( ) ( ) | | | |
( )m N / k 2 j DFT
N
DFT
e k X 1 - N n 0 m n x
k X n x
t
s s

| | | |
| | ( ) ( ) | |
N
DFT
DFT
k Nx n X
k X n x


Spring 2008 11
Example: Duality
Spring 2008 12
Symmetry Properties
Spring 2008 13
Circular Convolution
Circular convolution of of two finite
length sequences




| | | | ( ) ( ) | |

=
=
1 N
0 m
N 2 1 3
m n x m x n x
| | | | ( ) ( ) | |

=
=
1 N
0 m
N 1 2 3
m n x m x n x
Spring 2008 14
Example
Circular convolution of two rectangular pulses L=N=6



DFT of each sequence



Multiplication of DFTs



And the inverse DFT
| | | |

s s
= =
else 0
1 L n 0 1
n x n x
2 1
| | | |

=
= = =

=
t

else 0
0 k N
e k X k X
1 N
0 n
kn
N
2
j
2 1
| | | | | |

=
= =
else 0
0 k N
k X k X k X
2
2 1 3
| |

s s
=
else 0
1 N n 0 N
n x
3
Spring 2008 15
Example
We can augment zeros to
each sequence L=2N=12
The DFT of each sequence




Multiplication of DFTs

| | | |
N
k 2
j
N
Lk 2
j
2 1
e 1
e 1
k X k X
t

= =
| |
2
N
k 2
j
N
Lk 2
j
3
e 1
e 1
k X
|
|
|
.
|

\
|

=
t