UNIT II

DISCRETE FOURIER SERIES & FOURIER TRANSFORMS

Introduction
Fourier analysis is a family of mathematical techniques, all based on decomposing signals in to
sinusoidal. i.e., signal approximation using Sin and Cos signals which are mutually orthogonal.

Discrete Fourier series,

Properties of discrete Fourier series
DFS representation of periodic sequences,

DTFT
Discrete Fourier transforms:
Note:

• The DFT provides uniformly spaced samples of the Discrete-Time Fourier Transform
(DTFT)
DFT definition:
2nk
1 N 1
x[n]   X [k ]e
j
N
N 1 2nk
j
X [k ]   x[n]e N N n 0
n 0
Requires N2 complex multiplies and N(N-1) complex additions

Find DFT for the following sequence.

1)x(n) ={0,1}, 2) x(n)={1,1,0,0}, 3) x(n)={1,2,3,4,5,6,7,8}

1)x(n) ={0,1}
a)

X(k)={1,-1}
3) x(n)={1,2,3,4,5,6,7,8}
Note:
For a sequence of length N, the DFT equations are:

X(0)=x(0) . 1+ x(1). 1+ x(2) . 1+ x(3) . 1+ ---------------- x(N-1) . 1

X(1)=x(0) . WN0.1 + x(1). WN1.1 + x(2) . WN2.1 + x(3) . W N3.1 + ---------------- x(N-1) . WN( N 1).1
X(2)=x(0) . WN0.2 + x(1). WN1.2 + x(2) . WN2.2 + x(3) . WN3.2 + ---------------- x(N-1) . WN( N 1).2
:
:
:
:
X(N-1)=x(0) . WN0.( N 1) + x(1). WN1.( N 1) + x(2) . WN2.( N 1) + x(3) . WN3.( N 1) + ------- x(N-1) . WN( N 1).( N 1)

The above set of equations in the matrix form

 X (0)  1 1 1 1 . . 1   x(0) 
 X (1)  1 W 1 WN 2
W N3 . . W ( N 1)   x(1) 
   N N   
 X (2)  1 W N2 W N4 W N6 . . W 2.( N 1)
N
  x(2) 
     
 .   . . . . . . .   . 
 .  . . . . . . .   . 
     
 .  . . . . . . .   . 
 X ( N  1) 1 W ( N 1) W 2.( N 1) W 3.( N 1) W N( N 1)( N 1)   x( N  1)
   N N N . .  

X ( K )  [WNnk ] x(n)

If N=4

 X (0)  1 1 1 1   x(0)
 X (1)  1 W 1 W 2 W 3   x(1) 
  4 4 4   
 X (2) 1 W42 W44 W46   x(2)
   3 6 9  
 X (3)  1 W4 W4 W4   x(3) 

j 2
 2 2  
W4  e 4
 cos( )  j sin( )  cos( )  j sin( )  0  j 1   j
4 4 2 2
(W4)1 = (-j)1 = - j
(W4)2 = (-j)2 = - 1
(W4)3 = (-j)3 = j

(W4)4 = (-j)4 = 1

(W4)5 = (-j)5 = -j

(W4)6 = (-j)6 = -1

(W4)7 = (-j)7 = j

(W4)9 = (-j)9 = -j
1

W40

j W43 -j
W41

W42

-1

 X (0)  1 1 1 1   x(0)
 X (1)  1  j  1 j   x(1) 
    
 X (2) 1  1 1  1   x(2)
     
 X (3)  1 j  1  j   x(3) 

Ex: Find DFT x(n)={0,1,2,3}.

x(0)=0, x(1)=1, x(2)=2, x(3)=3

N 1
X ( K )   x(n) WNnk
n 0

X ( K )  [WNnk ] x(n)
 X (0)  1 1 1 1  0 
 X (1)  1  j  1 j  1
    
 X (2) 1  1 1  1  2
     
 X (3)  1 j  1  j  3

 X (0)   0 1 2  3 
 X (1)  0  j  2  3 j 
   
 X (2)  0 1 2  3 
   
 X (3)  0  j  2  3 j 

 X (0)   6 
 X (1)   2  2 j 
   
 X (2)  2 
   
 X (3)   2  2 j 

Therefore X(K)={6, -2+2j, -2, -2-2j}

If N=8
 X (0)  1 1 1 1 1 1 1 1   x(0) 
 X (1)  1 W 1
W 2
W 3
W 4
W 5
W 6
W87   x(1) 
   8 8 8 8 8 8  
 X (2) 1 W 8
2
W 8
4
W
8
6
W8
8
W
8
10
W8
12
W814   x(2)
   3 6 9 12 15 18   
 X (3)   1 W 8 W 8 W
8 W
8 W
8 W8 W821   x(3) 
 X (4) 1 W 4
W 8
W12
W16
W20
W 24
W828   x(4)
     
8 8 8 8 8 8
5 10 15 20 25 30
 X (5)  1 W 8 W 8 W
8 W
8 W
8 W8 W835   x(5) 
 X (6)  1 W 6
W 12
W18
W24
W30
W 36
W842   x(6) 
   8 8 8 8 8 8
  
 X (7) 1 W849   x(7)
7 14 21 28 35 42
W 8 W 8 W
8 W
8 W
8 W8

W80  1

j 2
 2 2   1 j 1 j
W e
8
1 8
 cos( )  j sin( )  cos( )  j sin( )   
8 8 4 4 2 2 2

j 2 2
 2 2 2 2  
W82  e 8
 cos( )  j sin( )  cos( )  j sin( )   j
8 8 2 2

j 2 3
 23 23 3 3 1 1 1 j
W e
8
3 8
 cos( )  j sin( )  cos( )  j sin( )  j 
8 8 4 4 2 2 2
 1 j 1
2 1 j
2
W87 W80
W81
j W86 -j
W82

W85 1 j
W83
1 j W84 2
2
-1

EX: Find DFT of x(n)={1,2,3,4,5,6,7,8}

1 1 1 1 1 1 1 1 
 X (0) 1 1 j 1 j 1 j 1  j   x(0)
j 1 j
 X (1)   2 2 2 2   x(1) 
  1  j 1 j 1  j 1 j   
 X (2)  1 j 1 j 1 j  1  j   x(2)
  1 j 1 j  
 X (3)    2 2 2 2   x(3) 
 X (4) 1 1 1 1 1 1 1  1   x(4)
   1 j 1 j 1 j 1 j   
 X (5)  1 j 1 j   x(5) 
 X (6)  2 2 2 2   
  1 j 1  j 1 j 1  j   x(6)
 X (7) 1 1 j 1 j
1
1 j
j
1  j   x(7)
 j 
 2 2 2 2 
X[K]=DFT(x(n))=
1 1 1 1 1 1 1 1 
 X (0)  1 1 j 1 j 1 j 1  j  1 
 j 1 j
 X (1)   2 2 2 2   2
  1  j 1 j 1  j 1 j   
 X (2)  1 j 1 j 1 j  1  j   3
  1 j 1  j  
 X (3)  2 2 2 2   4
 X (4) 1 1 1 1 1 1 1  1  5 
   1 j 1 j 1 j 1 j   
 X (5)  1  j 1 j  6 
 X (6)   2 2 2 2   
  1 j 1  j 1 j 1  j  7 
 X (7) 1 1 j 1 j
1
1 j
 j
1  j  8 
 j 
 2 2 2 2 
Properties of DFT,
Linear filtering methods based on DFT,
DFT
x(n)  X(K)

Fast Fourier transforms (FFT) : Radix-2 : Decimation in time (DIT)

and Decimation in frequency (DIF) FFT Algorithms:

 FFT is not a transform, it is an algorithm used to compute DFT efficiency.

Decimation In Time

x(n)={x(0), x(1), x(2), x(3), x(4), x(5), x(6), x(7)}

Decimation In Frequency

X(K)={X(0), X(1), X(2), X(3), X(4), X(5), X(6), X(7)}

Flow Graph of 8-point Decimation in Time FFT algorithm:

Construction of 8-point DFT from four 2-point DFTs

Construction of 8-point DFT from four 4-point DFTs
 j 2 j
   
W8  e 8
e 4
 cos( )  j sin( )
4 4
1 1
  j  0.707  j 0.707
2 2
 Stage -2 Stage -3
Stage -1 X(0)= S2(0) + W80 S2(4)
S1(0)= x(0)+W80 x(4) S2(0)= S1(0) + W80 S1(2)
x(0)
+ X(1) S1(0) + W80 S1(2)
S2(1)= S1(1) + W82 S1(3)
- S1(1)=x(0) – W8 x(4)
0
x(4) W80
W80 X(2)
S2(2)=S1(0) –W80 S1(2)
x(2) S1(2)=x(2)+W80 X(6)
+ W82 X(3)
S1(3)=x(2) - W80 X(6)W S2(3)
x(n) x(6) W80
- 8
0

W80
S1(4) S2(4) X(4) X(K)
x(1)
+
W81
- S1(5) S2(5)=S1(5) +W82 S1(7) X(5)
x(5) W80
S1(6) W80 S2(6) W82
X(6)
x(3)
+
S1(7) W82 W83 X(7)
x(7) W80 -
S2(7)=S1(5) - W82 S1(7)

X(0) S1(0) = x(0)+W80 x(4)=1/2+0=1/2

+

W80
X(4) - S1(1)=x(0) - W80 x(4)=1/2-0=1/2

X(2) S1(2) = x(2) + W80 x(6)=1/2+0=1/2

+

X(6) W80 - S1(3)=x(2) - W80 x(6)=1/2-0=1/2

Problem: Find Radix-2 Decimation In Time FFT of x(n)={1,2,3,4,1,2,3,4}
N=8, Number of stages=3, W80=1, W81= 0.707-j0.707, W82= -j, W83= -0.707-j0.707

Stage -2 Stage -3
Stage -1 X(0)= S2(0) + W80 S2(4)
S1(0)= x(0)+W80 x(4)= S2(0)= S1(0) + W80 S1(2)
x(0)=1
+ X(1) S1(0) + W80 S1(2)
S2(1)= S1(1) + W82 S1(3)
- S1(1)=x(0) – W8 x(4)
0
x(4)=1 W80
W80 X(2)
S2(2)=S1(0) –W80 S1(2)
x(2)=3 S1(2)=x(2)+W80 X(6)
+ W82 X(3)
S1(3)=x(2) - W80 X(6)W S2(3)
x(n) x(6)=3 W80 - 8
0

W80
S1(4) S2(4) X(4) X(K)
x(1)=2
+
W81
- S1(5) S2(5)=S1(5) +W82 S1(7) X(5)
x(5)=2 W80
S1(6) W80 S2(6) W82
X(6)
x(3)=4
+
S1(7) W82 W83 X(7)
x(7)=4 W80 -
2
S2(7)=S1(5) - W8 S1(7)

X(K)={20, 0, 4j, 0, -4, 0, -4j, 0}

 Stage -2 Stage -3
Stage -1 S1(0)=x(0)+x(4)=5 S2(0)=10 X(0)=20
x(0)=1
W80
W80 S1(1)=x(1)+x(5)=5 W80 S2(1)=10
X(4)=0
x(1)=2
S2(2)=0
S1(2)=x(2)+x(6)=5 X(2)=0
W82
W81
x(2)=3 W80
S1(3)=5 S2(3)=0
W82 X(6)=0
x(3)=4 X(K)
x(4)=4 W83
S1(4)=[x(0)-x(4)]W80= - 3 S2(4)=-3-j X(1)=-5.82-j2.414
x(n)
W80
1
x(5)=3 S1(5)=[x(1)-x(5)]W8 = -0.707+j 0.707 W80 S2(5)=-2.82-j1.414
X(5)=-0.18+j0.414

S2(6)=-3+j
x(6)=2 S1(6)=[x(2)-x(6)]W82=-j X(3)=5.82-j0.414
W82
W80
S2(7)=2.82-j1.414
x(7)=1 X(7)=0.18+j2.41
S1(7)=-2.12-j2.12 4
S1(0)=X*(0)+x*(4)W80=7+1=8

S1(1)=X*(0)-x*(4)W80=7-1=6