Digital Signal Processing Lab Manual

Digital Signal Processing Lab
2014
LAB MANNUAL
DIGITAL SIGNAL PROCESSING (EC-603)
Submit By
Name-..
Roll no-
Session -.
Submit to
Prof.
Prof. Prateek Mishra
Asst. professor
HOD
EC Dept. GNCSGI
EC Dept. GNCSGI
DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING, GNCSGI
GNCSGI Jabalpur
Page 1
2014
GLOBAL NATURE CARE SANGATHANS GROUP

OF INSTITUTION, JABALAPUR
Digital signal processing
(EC- 603)
List of Experiments
1. To plot and demonstrate various types of discrete time signals.
2. To plot and demonstrate shifting of discrete time signals.
3. To plot and demonstrate addition, subtraction multiplication of
discrete time signals.
4. To find the impulse response of a discrete time system described
by difference equation and also find its stability.
5. Implement an M-Point moving average filter.
6. Implementation of an integrator and differentiator.
7. To Sketch Pole zero plot of a given transfer function.
8. Finding the Discrete time Fourier Transform of a given impulse
response and verify its convolution property.
9. To verifying the convolution property of DTFT.
10. To implement FIR Low Pass Filter Using Windowing
11. To implement FIR High Pass Filter Using Windowing
GNCSGI Jabalpur
Page 2
2014
EC 603 Digital Signal Processing

Unit I
Discrete-Time Signals and Systems Discrete-time signals, discretetime systems, analysis of discrete-time linear time-invariant
systems, discrete time systems described by difference equation,
solution of difference equation, implementation of discrete-time
systems, stability and causality, frequency domain representation
of discrete time signals and systems.
Unit - II
The z-Transform The direct z-transform, properties of the
transform, rational z-transforms, inversion of the z transform,
analysis of linear time-invariant systems in the z- domain, block
diagrams and signal flow graph representation of digital network,
matrix representation.
Unit - III
Frequency Analysis of Discrete Time Signals Discrete fourie series
(DFS), properties of the DFS, discrete Fourier transforms (DFT)
properties of DFT, two dimensional DFT, circular convolution.
Unit - IV
Efficient Computation of the DFT- FFT algorithms, decimation in
time algorithm, decimation in frequency algorithm, decomposition
for N composite number.
Unit - V
Digital filters Design Techniques Design of IIR and FIR digital
filters, Impulse invariant and bilinear transformation, windowing
techniques- rectangular and other windows, examples of FIR
filters, design using windowing.
GNCSGI Jabalpur
Page 3
2014
Experiment detail
S no.
Object
To plot and demonstrate various types of discrete time

signals.
To plot and demonstrate shifting of discrete time signals.
To plot and demonstrate addition, subtraction

multiplication of discrete time signals.
To find the impulse response of a discrete time system

described by difference equation and also find its stability.
5
6
7
8
9
Pg no
Implement an M-Point moving average filter.

Implementation of an integrator and differentiator.
To Sketch Pole zero plot of a given transfer function.
Finding the Discrete time Fourier Transform of a given
impulse response and verify its convolution property.
To verifying the convolution property of DTFT.
10
To implement FIR Low Pass Filter Using Windowing
11
To implement FIR High Pass Filter Using Windowing
GNCSGI Jabalpur
Page 4
2014
1. Demonstration of basic discrete time signals.

Aim:- To plot and demonstrate various types of discrete time signals.
Equipments:Operating System Windows XP or above version
Constructor script or Simulator
Software - MATLAB 7 or above version
Theory:-
We use several elementary sequences in digital signal processing for analysis

purposes. Their definitions and MATLAB representations are given below.
1. Unit sample sequence:-
(n) = {
= {..,0,0,0,1,0,0,0,..
In MATLAB t he function zeros (1, N) generates a row vector of N zeros, which can be used to
implement (n) over a finite interval. However, the implementing (n) can be done by taking
n=0 sample equal to 1.
Program:% Digital Signal Processing Lab

% Program to generate a unit impulse signal
N=64;
n=-(N/2): ((N/2)-1);
x=zeros(1,N);
x((N/2)+1)=1;
stem(n,x);
grid;
title('A unit impulse signal ');
xlabel('Sampel Number');
ylabel('Amplitude');
GNCSGI Jabalpur
Page 5
2014
2. Unit step sequence:u(n) = {

In MATLAB t he function ones (1, N) generates a row vector of N ones. It can be used to
generate u(n) over a finite interval.

% Program to generate a unit step signal
n=-20:20;
u=[zeros(1,20),ones(1,21)];
stem(n,u);
grid;
title('A unit step signal ');
GNCSGI Jabalpur
Page 6
2014
3. Real-valued exponential sequence:x(n) = ()n ,

In MATLAB an array operator " . ^ " is required to implement a real exponential sequence.
For example, to generate z(n) = (0.9)n , 0 < n < 10, we will need the following MATLAB script:
>> n=0:10; x=(0.9).^n;

% Program to generate a real valued exponential signal
% x(n)=(0.9).^n
n=0:40;
x=(0.9).^n;
stem(n,x);
grid;
title('A Real valued Exponential signal ');
4. Sinusoidal sequence:x(n) = cos (o n + )

where o = 2 /N where N is period of this discrete time
sinusoid. Where is the phase in radians, a MATLAB function cos (or sin) is used to generate
sinusoidal sequences.
GNCSGI Jabalpur
Page 7
2014

% Program to generate a sinusoidal signal
% x(n)=A*cos((2*pi/N)+ theta)in this example N=12 and theta is pi/3
n=0:40;
x=2*cos((2*(pi/12)*n)+pi/3);
stem(n,x);
grid;
title('A sinusoidal signal of period N=12');
xlabel('Sample Number');
GNCSGI Jabalpur
Page 8
2014
2. Demonstration of shifting of discrete time signals.

Aim:- To plot and demonstrate shifting of discrete time signals.
Equipments:Operating System Windows XP
Constructor - Simulator
Software - MATLAB 7
Theory:-
Shifting is any discrete time sequence can be seen as delaying or advancing the
sequence in time or sample interval if x(n) is a discrete time signal than its delayed version can
be expressed as x(n- no) where no is amount of shift. To implement this using MATLAB we have
to pad no numbers of zeros before starting the sequence x(n).

% Program to generate a sinusoidal signal and delay it by 10 samples
% x(n)=A*cos((2*pi/N)+ theta)in this example N=12 and theta is 0 and
% amount of delay is of 20 samples
N=64
n=0:N-1;
x=2*cos(2*(pi/12)*n);
xd =[zeros(1,20), x(1:(N-20))];
subplot(2,1,1);
stem(n,x);
grid;
title('A sinusoidal signal of period N=12');
subplot(2,1,2);
stem(n,xd);
grid;
title('A delayed sinusoidal signal of period N=12 amount of delay =20');
% in this section we will generate a delayed unit impulse sequence
% delayed by 20 samples
m=-(N/2): ((N/2)-1);
u=zeros(1,N);
u((N/2)+1)=1;
ud=[zeros(1,20),u(1:N-20)];
figure(2)
subplot(1,2,1)
stem(m,u);
GNCSGI Jabalpur
Page 9
2014
grid;
title('A unit impulse signal ');
subplot(1,2,2)
stem(m,ud);
grid;
title('A unit impulse signal delayed by 20 samples');
GNCSGI Jabalpur
Page 10
2014
3. Demonstration basic manipulation on discrete time signals.
Aim:- To plot and demonstrate addition, subtraction multiplication of discrete

time signals.

Software - MATLAB 7
Theory:1. Signal addition: - This is a sample -by-sample addition given by x1(n) + x2(n) it is
implemented in MATLAB by the arithmetic operator +. However, the lengths of x 1(n)

and x2(n) must be the same.
2. Signal subtraction:- This is a sample -by-sample subtraction given by x1(n) - x2(n)
It is implemented in MATLAB by the arithmetic operator -. However, the lengths of
x1(n) and x2(n) must be the same.
3. Signal subtraction:- This is a sample -by-sample multiplication given by x1(n) *x2(n)
It is implemented in MATLAB by the arithmetic operator *. However, the lengths of
x1(n) and x2(n) must be the same. This operation is non linear operation whereas addition
and subtraction are linear operations.
Program :% Digital Signal Processing Lab

% Program to demonstrate addition subtraction and multiplication
% of two different sinusoids having periods N1=15 and N2=30
n=0:50;
x1=cos(2*pi/15*n);
x2=0.5*sin(2*pi/30*n);
add=x1+x2;
sub=x1-x2;
mul=x1 .* x2;
subplot(1,2,1);
stem(n,x1);
grid;
title('A sinusoidal signal of N=15');
subplot(1,2,2);
stem(n,x2);
GNCSGI Jabalpur
Page 11
2014
grid;
title('A sinusoidal signal of N = 30');
figure(2);
subplot(1,3,1);
stem(n,add);
grid;
title('A addition x1+x2');
subplot(1,3,2);
stem(n,sub);
grid;
title('A subtraction x1-x2');
subplot(1,3,3);
stem(n,mul);
grid;
title('A multiplication x1*x2');
GNCSGI Jabalpur
Page 12
GNCSGI Jabalpur
2014
Page 13
2014
4. Finding impulse response of a given discrete time system and
finding its stability.

Aim:- To find the impulse response of a discrete time system described by
difference equation and also find its stability.

Software - MATLAB 7
Theory:-
Impulse response of a given system described by a difference equation can be

found by using MATLAB function impz(num,den,N) where num is a row vector of coefficient
of input x(n) = (n) and its delayed forms, similarly den is a row vector of coefficients of out
put y(n) and its delayed forms. And N is total number of samples for which we have to find
response.
Where as the test of stability of a system can be performed using the method Sum of
Absolute Test in this method after finding impulse response of a system for N samples we test
the absolute value of h(n) i.e. impulse response, if the absolute value is decreasing as N is
increasing and if the absolute value of h(n) becomes < 10-6 the system can be considered as
stable system.

% Program to find impulse response of a system
% y(n)+ 1.5 y(n-1)+ 0.9 y(n-2)= x(n)- 0.8 x(n-1)
clc;
N=200;
n=0:N;
num = [1 -0.8];
den = [1 1.5 0.9];
h = impz(num,den,N+1);
stem(n,h);
xlabel('Time index n');
title('Impulse Response h(n)');
sum = 0;
for K=1:N+1;
sum = sum + abs(h(K));
if abs(h(K))< 10^-6;
break
end
end
stem(n,h);
GNCSGI Jabalpur
Page 14
2014
xlabel('Time index n');

title('Impulse Response h(n)');
grid;
disp('value = ');
disp(abs(h(K)));
disp('sum of h(n) from n = 0 to N = ')
disp(sum);
Out Put:value =
1.6761e-0.05
sum of h(n) from n = 0 to N =
35.3591
GNCSGI Jabalpur
Page 15
2014
5. M-Point Moving Average Filter
Aim:- Implement an M-Point moving average filter.

Software - MATLAB 7
Theory: - an M pint moving average filter can be described as below.

( )
This can be implemented using MATLAB function filter and the matlab script to implement
this can be written as: >> y=filter(num,1,x);
Where x(n) is input and num is a row vector of all the elements 1 and it is a 1xM matrix, and 1 is
denominator co efficient.

% Program to impliment M point moving average filter
% y(n)= 1/M(x(n)+ x(n-1)+x(n-2)+ x(n-3)+x(n-4)+ ......+x(n-M))
close all;
clear all;
clc;
clf;
n= 0:56;
s1=cos(2*pi*0.05*n);
s2=cos(2*pi*0.5*n);
s=s1+s2;
subplot(2,2,1);
plot(n,s1);
grid;
title('Orignal signal s1 ,N=20 ');
xlabel('number of sampel');
subplot(2,2,2);
plot(n,s2);
grid;
title('Additive Noise signal s2 ,N=2 ');
subplot(2,2,3);
plot(n,s);
grid;
title('Noisy signal to be filtered = s1+s2 ');
GNCSGI Jabalpur
Page 16
2014
M=input('Please enter the length of filter = ');
num=ones(1,M);
y=filter(num,1,s)/M;
subplot(2,2,4);
plot(n,y);
grid;
title(' Filtered signal ');
Please enter the length of filter = 10
GNCSGI Jabalpur
Page 17
2014
6. Implementation of integrator and differentiator
Aim:- implementation of an integrator and differentiator.

Software - MATLAB 7
Theory: - In digital signal processing a differentiator and integrator can be synthesized using
difference equations,
1. Differentiator:- a differentiator can be implemented using finding the first difference of
any given input signal x(n) and it is given as:
y(n) = x(n)-x(n-1) this is called the first difference of input x(n).
2. Integrator or Accumulator:- An integrator can be viewed as :
y(n)=
( ) and this can be converted in to difference equation as:
simulation of Integration Using Trapezoidal rule y(n)=y(n-1)+1/2(x(n)+x(n-1))

% Program to find integration and differentiation of a given signal x(n)
% the integrator can be implemented using y(n)- y(n-1)= x(n)
% and differentiation can be implemented using y(n)= x(n)- x(n-1)
% in this program we will use x(n)= u(n)i.e. unit step signal
n=-50:50;
u=[zeros(1,50) ones(1,51)];
axis([-50,50,0,1]);
subplot(3,1,1);
stem(n,u);
grid;
title('imput unit step signal');
xlabel('sampel number');
% implementing differentiation y(n)= x(n)- x(n-1)
num1=[1,-1];
den1=[1];
yd=filter(num1,den1,u);
subplot(3,1,2);
stem(n,yd);
grid;
title('Differentiation of unit step signal is impulse signal');
% implementing integration y(n)- y(n-1)= x(n)+ x(n-1)
num2=[1/2,1/2];
GNCSGI Jabalpur
Page 18
2014
den2=[1 ,-1];
yi=filter(num2,den2,u);
subplot(3,1,3);
stem(n,yi);
grid;
title('Integration of unit step signal is unit ramp signal');
GNCSGI Jabalpur
Page 19
2014
7. Z transform and pole zero plot
Aim:- Sketching Pole zero plot of a given transfer function.

Software - MATLAB 7
Theory:- Assume we have a transfer function in the z-domain given by:X z
z 1
z
2
1
2
1 0.25 z 0.375 z
z 0.25 z 0.375
Factoring this, its partial fraction expansion can be found to be:c1 z

c2 z
z
0.8 z
0.8 z
X z
z 0.75 z 0.5 z 0.75 z 0.5 z 0.75 z 0.5

The inverse z-transform of this is thus:-
xn 0.80.75 0.8 0.5 u n

n
There are several MATLAB functions that could assist with calculating and analyzing these
results. We can find the roots of the denominator polynomial using,
den = [1 -0.25 -0.375]; roots(den)
We can then plot the zeros and poles either with
1. Zeros and poles in column vectors
zplane(z,p) z = [0] p = [0.75; -0.5]
2. numerator and denominator coefficients in row vectors
num = [0 1 0]
den = [1 -0.25 -0.375]
zplane(num,den)
GNCSGI Jabalpur
Page 20
2014
Program:-
% Digital Signal Processing Lab

% Program to show pole zero location in complex z plane
% for a given transfer function,Described in theory
clc;
close all;
clear all;
num=[0];
den=[1,-0.25,-0.375];
z=roots(num);
p=roots(den);
% Pole zero plot using Zeros and poles in column vectors
zplane(z,p);
title('Pole zero plot using Zeros and poles in column vectors');
figure(2);
%Pole zero plot using numerator and denominator coefficients in row vectors
zplane(num,den);
title('Pole zero plot using numerator and denominator coefficients in row vectors');
GNCSGI Jabalpur
Page 21
GNCSGI Jabalpur
2014
Page 22
2014
8. Discrete time Fourier Transform and its Convolution
property
Aim: - Finding the Discrete time Fourier Transform of a given impulse response
and verify its convolution property.

Software - MATLAB 7
Theory:- As we know that DTFT of a discrete time signal is continuous and

periodic in nature having period 2 . In matlab using function freqz we can
compute H(ei) i.e. the Fourier transform of any given system described by
difference equation , and we can plot its real value and imaginary value and with
respect to frequency in radian and we can also plot the magnitude spectrum and
phase spectrum of that system.
Program:% Digital signal processing Lab

% Program to find DTFT of a given system described by difference equation
% y(n)- 0.6 y(n-1)= 2 x(n)+ x(n-1)
close all;
clear all;
clc;
clf;
% computing the frequency samples of the DTFT
w=-4*pi:8*pi/511:4*pi;
num=[2 1]; den=[1 -0.6];
H=freqz(num,den,w);
% plot the DTFT
subplot(2,1,1);
plot(w/pi, real(H)); grid;
title('Real part of H(e^{j\omega})');
xlabel('\omega/\Pi');
subplot(2,1,2);
GNCSGI Jabalpur
Page 23
2014
plot(w/pi, imag(H)); grid;

title('Imaginary part of H(e^{j\omega})');
figure(2);
subplot(2,1,1);
plot(w/pi, abs(H)); grid;
title('Magnitude Spectrum of H(e^{j\omega})');
subplot(2,1,2);
plot(w/pi, angle(H)); grid;
title('Phase Spectrum of H(e^{j\omega})');
ylabel('Phase in Radians');
Out Put:-
GNCSGI Jabalpur
Page 24
GNCSGI Jabalpur
2014
Page 25
2014
9. convolution property of DTFT.
Aim: - verifying the convolution property of DTFT.

Software - MATLAB 7
Theory:- As we know that if two sequences are convolved in time

domain than their frequency responses i.e. DTFT multiplies in frequency
domain to demonstrate this property we first convolve two sequences
and will find Fourier transform of convolution, on the other side we
separately find the discrete time Fourier transform of two sequences and
then multiply both of them to prove the convolution property of DTFT.
Basic Matlab function used will be the same i.e. freqz.
Program:% Digital signal processing Lab
% Demonstration of convolution property of DTFT
% x1 = [1 3 5 7 9 11 13 15 17]
% x2 = [1 -2 3 -2 1]
close all;
clear all;
clc;
clf;
w= -pi:2*pi/255:pi;
x1 = [1 3 5 7 9 11 13 15 17];
x2 = [1 -2 3 -2 1];
y = conv(x1,x2);
H1 = freqz(x1,1,w);
H2 = freqz(x2,1,w);
Hp = H1.*H2;
H3 = freqz(y,1,w);
subplot(2,2,1);
plot(w/pi,abs(Hp));
grid;
title('Magnitude spectrum of convolved sequence Hp(e^{j\omega})');
GNCSGI Jabalpur
Page 26
2014
subplot(2,2,2);
plot(w/pi,abs(H3));
grid;
title('Product of magnitude spectrum H3(e^{j\omega})');
subplot(2,2,3);
plot(w/pi,angle(Hp));
grid;
title('Phase spectrum of convolved sequence Hp(e^{j\omega})');
ylabel('Angle in Radian');
subplot(2,2,4);
plot(w/pi,angle(H3));
grid;
title('Phase spectrum of Product of Phase spectrum H3(e^{j\omega})');
ylabel('Angle in Radian');
Out Put:-
GNCSGI Jabalpur
Page 27
2014
10. Designing of Fir Low Pass Filter
Aim:- To implement FIR Low Pass Filter Using Windowing

Theory:Desired magnitude response of the ideal filter is given by the equation

(
Where is shift =(N-1)/2 where N= length of window (odd)

And c = cut off frequency of LPF
|H(ej)
|
Figure 1 Frequency response of FIR low pass filter with cutoff frequency c
Corresponding impulse response of LPF will be:

( )
(
(
)
)
GNCSGI Jabalpur
Page 28
2014
( )
Where N is length of window having odd length
Program:clc;
clf;
close all;
clear all;
clc;
% designing of LPF using Hamming window
% take cut off frequency wp as input
% h(n)= sin[wp(n-t)]/pi*(n-t) when n != t
% h(n)= wp/pi
when n=t
f=input('Give cutoff frequency in hertz between 0 to 4kHZ
wp=pi*f/4000;
N = 1025;
t=(N-1)/2;
n=0:N-1;
h1=sin(wp*(n-t));
h2=(pi*(n-t));
h1((n-t)==0)=wp;
h2((n-t)==0)=pi;
h=h1./h2;
stem(n,h);
title('impulse response of filter');
w = 0.54-0.46*(cos(2*pi*n/(N-1)));
figure(2);
stem(n,w);
title('Hamming window');
y=h.*w;
figure(3);
stem(n,y);
title('impulse response of filter h(n)*w(n)');
z=fftshift(abs(fft(y)));
f=-4000:(8000/1024):4000;
figure(4)
plot(f,z);
xlabel('frequency in hertz');
GNCSGI Jabalpur
');
Page 29
2014
ylabel('Amplitude Spectrum');
title('Amplitude spectrum of FIR LPF');
Out Put:-
GNCSGI Jabalpur
Page 30
2014
Aim:- To implement FIR High Pass Filter Using Windowing

Theory:Desired magnitude response of the ideal filter is given by the equation

(
)
(
Where is shift =(N-1)/2 where N= length of window

And c = cut off frequency of HPF
(
Corresponding impulse response of LPF will be:

( )
(
(
)
)
( )
Where N is length of window having odd length
GNCSGI Jabalpur
Page 31
2014
Program:clc;
clf;
close all;
clear all;
clc;
% designing of HPF using Hamming window
% take cut off frequency wp as input
% h(n)= -sin[wc(n-t)]/pi*(n-t) when n != t
% h(n)= pi-wc/pi
when n=t
f=input('Give cutoff frequency in hertz between 0 to 4kHZ
wc=pi*f/4000;
N = 1025;
t=(N-1)/2;
n=0:N-1;
h1=-sin(wc*(n-t));
h2=(pi*(n-t));
h1((n-t)==0)=pi-wc;
h2((n-t)==0)=pi;
h=h1./h2;
stem(n,h);
title('impulse response of filter');
w = 0.54-0.46*(cos(2*pi*n/(N-1)));
figure(2);
stem(n,w);
title('Hamming window');
y=h.*w;
figure(3);
stem(n,y);
title('impulse response of filter h(n)*w(n)');
z=fftshift(abs(fft(y)));
f=-4000:(8000/1024):4000;
figure(4)
plot(f,z);
xlabel('frequency in hertz');
ylabel('Amplitude Spectrum');
title('Amplitude spectrum of FIR HPF');
GNCSGI Jabalpur
');
Page 32
2014
Results:
GNCSGI Jabalpur
Page 33

Digital Signal Processing Lab Manual

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Digital Signal Processing Lab Manual

Uploaded by

Copyright:

Available Formats

Digital Signal Processing Lab

Prof. Prateek Mishra

DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING, GNCSGI