DIGITAL SIGNAL

PROCESSING
(DSP)
FUNDAMENTAL
S
Dr. Bhagya H K
Professor
Dept of ECE
KVGCE, Sullia.
 Digital Signal
• Converting a continuously changing
waveform (analog) into a series of discrete
levels (digital)
 Digital Signal
• The analog waveform is sliced into equal
segments and the waveform amplitude is
measured in the middle of each segment
• The collection of measurements make up
the digital representation of the waveform
What is DSP?
 SYLLABUS

Module 1:
a)Basic of Discrete Fourier Transform(DFT)
b)DFT and IDFT for N=4,8
c)Properties

Module 2:
a)Circular Convolution-problems and properties
b) Overlap-save and overlap-add method to find
DFT
Module 3:
Radix-2 FFT algorithm for the computation of DFT
IDFT–decimation-in-time and decimation-in-
frequency algorithms.
Goertzel algorithm, and chirp-z transform
Module 4:
a)Structure for IIR Systems: Direct form, Cascade
form, Parallel form structures.
b)Butterworth and Chebyshev filters
c)Analog to analog frequency transformations.
d) Design of IIR Filters from analog filter using
Butterworth filter: Impulse invariance, Bilinear
transformation.
Module 5:
Structure for FIR Systems: Direct form, Linear
Phase, Frequency sampling structure, Lattice structure.

FIR filter design: Introduction to FIR filters, design

of FIR filters using - Rectangular, Hamming, Hanning
and Bartlett windows.
 Module-2

The Fourier transform of discrete-time signals is called the

discrete-time Fourier transform (DTFT).

If x(n) is the given discrete-time sequence, then X(ω) is the

discrete-time Fourier transform of x(n).

The DTFT of x(n) is defined as:

The Discrete-time Fourier transform (DTFT) X(ω) of a discrete-
time sequence x(n) is a periodic continuous function of ω with a
period of 2π. So it cannot be processed by a digital system.

The DFT converts the continuous function of ω to a discrete

function of ω. Thus, DFT allows us to perform frequency analysis
on a digital computer.

The DFT is obtained by sampling one period of the Fourier

transform X(ω ) of the signal x(n) at a finite number of frequency
points. This sampling is conventionally performed at N equally
spaced points in the period 0 ≤ ω ≤ 2π or at k = 2 π k/N;
0 ≤ k ≤ N – 1.
Let x(n) be a discrete-time sequence with Fourier transform
X(ω), then the DFT of x(n) denoted by X(k) is defined as:
Compute the DFT of each of the following finite length
sequences considered to be of length N:
(a) x(n) = δ(n)

(b) x(n) = δ(n – n0), where 0 < n0 < N

(c) x(n) = an, 0 ≤n ≤ N – 1

{
(d) x(n) = 1 for even
0 for odd
(a) Find the 4-point DFT of x(n) = {1, –1, 2, –2} directly.

USE
 TRY

Solution:
 Note
DFT IDFT

[1 -1 2 -2] [0 -1-j 6 -1+j ]

[1 -2 3 2] [4 -2+4j 4 -2 -4j]

[1 1 0 0] [2 1-j 0 1+j ]

Complex Conjugate
Compute the DFT of the 3-point sequence x(n) = {2, 1, 2}.
Using the same sequence, compute the 6-point DFT and
compare the two DFTs.

The given 3-point sequence is x(n) = {2, 1, 2}, N = 3.

The computation procedure to find DFT formula is very
lengthy and cumbersome.
Instead we use the powers of WN in the multiplications
instead of e–j(2π/N) and its factors.
 TRY

X[n] X(k)

[0 -1-j 6 -1+j ]
[1 -1 2 -2]

[1 -2 3 2] [4 -2+4j 4 -2 -4j]

[2 1-j 0 1+j ]
[1 1 0 0]
 1) Periodicity
If a sequence x(n) is periodic with periodicity of N samples, then N-
point DFT of the sequence, X(k) is also periodic with periodicity of N
samples. Hence, if x(n) and X(k) are an N-point DFT pair, then

Proof: By definition of DFT, the (k + N)th coefficient of X(k) is given

by
2) Linearity
DFT of Even and Odd Sequence

Time Reversal of the sequence

The time reversal of an N-point sequence x(n) is obtained by

wrapping the sequence x(n) around the circle in the clockwise
direction. It is denoted as x[(–n), mod N] and
x[(–n), mod N] = x(N – n), 0 ≤n≤ N – 1
Circular Shift Property
Module-2
Graphical Method to perform circular convolution

In graphical method, also called concentric circle method, the given

sequences are represented on concentric circles. Given two sequences
x1(n) and x2(n) the circular convolution of these two sequences, x3(n)
= x1(n) x2(n) can be found using the following steps:

Step 1: Graph N samples of x1(n), as equally spaced points around an

outer circle in anticlockwise direction.

Step 2: Starting at the same point as x1(n), graph N samples of x2(n)

as equally spaced points around an inner circle in the clockwise
direction. This corresponds to x2(–n).
Step 3: Multiply the corresponding samples on the two circles
and sum the products to produce output,

Step 4: Rotate the inner circle one sample at a time in

anticlockwise direction and go to Step 3 to obtain the next
value of output. If it is rotated by q samples,

Step 5: Repeat Step 4 until the inner circle first sample lines up
with the first sample of the outer circle once again.
Relationship between DTFT and DFT

DTFT and
its Inverse
Let x(n) be a discrete-time sequence with Fourier transform
X(ω), then the DFT of x(n) denoted by X(k) is defined as:
Summary
: The DFT is obtained by sampling one period of the
Fourier transform X(ω ) of the signal x(n) at a finite
number of frequency points. This sampling is
conventionally performed at N equally spaced points in
the period 0 ≤ ω ≤ 2π or at k = 2 π k/N;
0 ≤ k ≤ N – 1.
Relationship between DTFT and Z-Transform
 FAST FOURIER TRANSFORM

In this chapter we look at the computational requirements

of the DFT and derive some fast algorithms for
computing the DFT. These algorithms are known,
generically, as Fast Fourier Transforms (FFTs).
1) Radix-2 decimation-in-time FFT, an algorithm
published in 1965 by Cooley and Tukey.
The N -point DFT of an N -point sequence s(n) is

Because x(n) may be either real or complex, evaluating X(k) requires on the
order of N complex multiplications and N complex additions for each value of
k. Therefore, because there are N values of X(k), computing an N-point DFT
requires N2 complex multiplications and additions.
The basic strategy that is used in the FFT algorithm is one of "divide
and conquer." which involves decomposing an N-point DFT into
successively smaller DFTs.

To see how this works, suppose that the length of x(n) is even (i.e.,
N is divisible by 2). If x(n) is decimated into two sequences of
length N/2, computing the N/2-point DFT of each of these
sequences requires approximately (N /2)2 multiplications and the
same number
of additions.

Thus, the two DFTs require 2 (N/ 2 )2= multiplies and adds.
Therefore, if it is possible to find the N-point DFT of s(n) from these
two N/2-point DFTS in fewer than N2/2 operations, a savings has
been realized.
The decimation-in-time FFT algorithm is based on splitting
(decimating) x(n) into smaller sequences and finding X ( k )
from the DFTs of these decimated sequences.

Let x(n) be a sequence of length N = 2v, and suppose that x(n) is

split (decimated) into two subsequences, each of length N/2. As
g(n) is, formed from the even-index terms,

and the second, h(n), is formed from the odd-index terms,

In terms of these sequences, the N -point DFT of x(n) is
 Note that the first term is the N/2-point DFT of g(n), and the
second is the N /2-point DFT of h(n):

Although the N/2-point DFTs of g(n) and h(n) are sequences of

length NL2, the periodicity of the complex exponentials allows us
to write
Therefore, X(k) may be computed from the N/2-point DFTs G
( k ) and H(k). Note that because

The basic computational unit of the FFT is called a

butterfly.
 Flow Diagram of an 8-point Decimation in time FFT

Position
Index Input x[n]

0 x(0) X(0)
W2 0
W8 0
W 40
x(4) X(1)
1 W21
W 41 W 81

2 x(2) X(2)
W2 0 W
4
2

W 82

3 x(6) X(3)
W21 W43 W83

X(4)
4 x(1) W 20
W 84
W85
W 40
X(5)
5 x(5) W21
W 41 W86

X(6)
6 x(3) W 4
2

W 20 W87
X(7)
x(7)
7 W21 W43

N=2 N=4
 Find the 8-point DFT of a sequence
x[n]=(1,1,1,1,0,0,0,0) using DIT-FFT radix-2
algorithm. Use Butterfly Diagram

Solution:

W 40 = 1 W80 = 1 W 84 = - 1
W20 =1
W41 = -j W81 = 0.707-j0.707 W85 = - 0.707+j0.707

W82 =-j W 86 =+j

W21 = -1 W4 =-1
2

W83 =- 0.707-j0.707 W87 = 0.707+j0.707

W 43 = j
Position
Index Input x[n]

0 1 1+0=1 1+1=2 2+2=4

1 1 1

1 0 1-0=1 1-j 1-j(2.41)

-1 -j
0.707-j0.707

2 1 1+0=1 1-1=0
1
-1 0
-j
0 1-0=1 1+j 1-j(0.414)
3 -1 j -1
-1
1 1+0=1 2-2=0
4 1+1=2
1 - 0.707+j0.707
1
5 0 1-0=1 1-j 1+j(0.414)
-1
-j j

6 1 1-1=0 0
1+0=1 -1
W 8
5

1 1+j(2.41)
1+j
7 0 1-0=1 j
-1
Further Reduction: Cooley Tukey Algorithm

WNr+N/2 = e –j(2π/N)(r+N/2) = e –j(2πr/N) e-j π =-WNr

a a a+b WNr
WNr
WN r
b
b a-b WNr
W Nr -1
-W N
r

a a+b WNr

W Nr
b a-b WNr
The number of complex multiplications required for calculating
N point DFT using DIT-FFT algorithm based on Cooley-Tukey is
(Nlog 2 N)/2. Also, since there are two additions per butterfly.

N/2 butterflies per stage and log2N stages.

The total number of complex additions needed is N log 2N

 Reduced Flow-graph for an 8-point DIT-FFT algorithm
stage
Position
Index Input x[n]
X(0)
0 x(0)

x(4) X(1)
1 W20 -1
x(2) X(2)
2 W40 -1
3 x(6) X(3)
W 20 -1 W4 1
-1 X(4)
4 x(1) -1
W80
5 x(5) X(5)
W20 -1 W81 -1
6 x(3) X(6)
W 40 -1 W8 2
-1
7 x(7) W 0 X(7)
2 -1 W4 1
-1 W8 3
-1
 Find the 8-point DFT of a sequence
x[n]=(1,1,1,1,0,0,0,0) using DIT-FFT radix-2
algorithm. Use Butterfly Diagram

Solution:

W 40 = 1 W80 = 1
W20 =1
W41 = -j W81 = 0.707-j0.707

W82 =-j
W21 = -1 W4 =-1
2

W83 =- 0.707-j0.707
W 43 = j
Input x[n]

x(0) 1 1+0=1 1+1=2

2+2=4

1-j
x(4) 0 1-0=1 1-j(2.41)
1 -1
1+0=1 1-1=0 0
x(2) 1
1 -1

1-0=1 1+j 1-j 0.414

x(6) 0 -j -1
1 -1

x(1) 1+1=2
1 1+0=1 2-2=0
-1
1
x(5) 1-j 1+j 0.414
0 1-0=1
-1
1 -1 0.707-j0.707
2-2=0
1-1=0 -1
x(3) 1 1+0=1 -1
-j
1
1+j 1+j(2.41)
1-0=1 -0.707-j0.707 -1
x(7) 0 -j -1
1 -1
 Find the 8-point DFT of a sequence
x[n]=(1,2,3,4,4,3,2,1) using DIT-FFT radix-2
algorithm. Use Butterfly Diagram

Solution:

W 40 = 1 W80 = 1 W 84 = - 1
W20 =1
W41 = -j W81 = 0.707-j0.707 W85 = - 0.707+j0.707

W82 =-j W 86 =+j

W21 = -1 W4 =-1
2

W83 =- 0.707-j0.707 W87 = 0.707+j0.707

W 43 = j
Input x[n]

x(0) 1 1+4=5 5+5=10

10+10=20

-3-j
x(4) 4 1-4=-3 -5.828-j2.414
1 -1
3+2=5 5-5=0 0
x(2) 3
1 -1

3-2=1 -3+j -0.171-j0.414

x(6) 2 -j -1
1 -1

x(1) 5+5=10
2 2+3=5 2-2=0
-1
1
x(5) -1-3j -0.171+0.414
3 2-3=-1
-1
1 -1 0.707-j0.707
2-2=0
5-5=0 -1
x(3) 4 4+1=5 -1
-j
1
-1+3j -5.828+j2.414
4-1=3 -0.707-j0.707 -1
x(7) 1 -j -1
1 -1
 Decimation-in-Frequency FFT
Another class of FFT algorithms may be derived by decimating the output
sequence X(k) into smaller and smaller subsequences. These algorithms are
called decimation-in-frequency FFTs and may be derived as follows.
Let N be a power of 2, N = 2v. and consider separately evaluating the even-index
and odd-index samples of X(k).
The even samples are

With a change in the indexing on the second sum we have

which is the N /2-point DFT of the sequence that is formed by adding the first N/2
points of x(n) to the last N/2.

Proceeding in the same way for the odd samples of X(k) leads to
 The basic butterfly for DIF-FFT
a a+b

b (a-b) WNr
-1
WN r
Input x[n] Output x[k]
x(0) X(0)
x(1) X(4)
-1 W20

x(2) X(2)
-1 W4 0
x(3) -1 W4
1 -1 W20 X(6)
x(4) X(1)
-1 W80
x(5) X(5)
-1 -1 W20
W81
x(6) X(3)
-1 W82 -1 W40
x(7) X(7)
-1 W83 -1 W41 -1 W0
 Find the 8-point DFT of a sequence
x[n]=(1,1,1,1,0,0,0,0) using DIF-FFT radix-2
algorithm. Use Butterfly Diagram

Solution:

W 40 = 1 W80 = 1 W 84 = - 1
W20 =1
W41 = -j W81 = 0.707-j0.707 W85 = - 0.707+j0.707

W82 =-j W 86 =+j

W21 = -1 W4 =-1
2

W83 =- 0.707-j0.707 W87 = 0.707+j0.707

W 43 = j
Input x[n]
1
x(0)=1 X(0)=4
2
1
x(1)=1 X(4)=0
2 -1 1
1 0
x(2)=1 X(2)=0
-1 1
1 0
x(3)=1 -1 1 X(6)=0
-1 -j

-1
x(4)=0
1 1
X(1)=1-j(2.414)
1-j
-1 0.707-j 0.707
x(5)=0 X(5)=1+j0.414
0.707-j 0.707 -j(1.414) -1 1

-1
-j 1+j
x(6)=0 X(3)=1-j(0.414)
-j -1 1
-1 -j(1.414)
x(7)=0 -0.707-j 0.707
-1 1 X(7)=1+j2.414
-0.707-j 0.707 -1 -j
 Find the 8-point DFT of a sequence
x[n]=(1,2,2,2,1,0,0,0) using DIF-FFT radix-2
algorithm. Use Butterfly Diagram

Solution:

X(0)=8
X(1)=-2j(2.414)
X(2)=0
X(3)=2j(0.414)
X(4)=0
X(5)= -2j(0.414)
X(6)=0
X(7)= 2j(2.414)
 Find the 4-point circular convolution of x[n]=(1,1,1,1)
and h(n)=[1,0,1,0] using DIF-FFT radix-2 algorithm.

Solution:

W 40 = 1
W20 =1
W41 = -j

W21 = -1 W42 =-1

W 43 = j
 Input x[n]

x(0)=1
X(0)=4
2
x(1)=1 X(2)=0
2 -1 1

0
x(2)=1 X(1)=0
-1 1 X(k)={4,0,0,0}
0
x(3)=1 X(3)=0
-1 -j -1 1
Input h[n]

h(0)=1
H(0)=2
2
h(1)=0
-1 1 H(2)=2 H(k)={2,0,2,0}
0
0
h(2)=1 H(1)=0
-1 1
0
h(3)=0 H(3)=0
-1 -j -1 1
 Y(k)=IFFT(X(k)H(k))=IFFT(8,0,0,0)
Input x[n]

Y(0)=8 2
8
8 1/4
Y(1)=0 8 2
0 -1 1
1/4
Y(2)=0
8
8 2
-1 1 1/4
0
Y(3)=0 8 2
-1 -j -1 1
1/4
First 5 points of the 8-point DFT of a real valued sequence is
given by X(0)=0 , X(1)=2+2j, X(2)= -4j , X(3)=2-2j , X(4)=0

Determine the remaining points. Hence find the original sequence

x(n) using DITFFT algorithm

Soln: X(5)=X *(3)=2+2j

X(6)= X *(2)=4j
X(7)= X *(1)=2-2j

*
Take Complex conjugates of Twiddle factors to
calculate IDFT
Input x[n]

X(0) 0 0 0
x(0)=8/8=1

8
X(4) 0 0 x(1)=8/8=1
1 -1
x(2)=-8/8=-1
X(2) -4j 0 0
1 -1

4j -8j -8 x(3)=-8/8=-1
X(6) j -1
1 -1

X(1) 2+2j 8
4+4j x(4)-8/8=-1
-1
1
X(5) 2+2j 0 0 x(5)=8/8=1
-1
1 -1 0.707+j0.707
x(6)=8/8=1
8j -1
X(3) 2-2j 4-4j -1
j
1
0 x(7)=-8/8=-1
2-2j 0 -0.707+j0.707 -1
X(7) j -1
1 -1