Scribd
Upload
67 views

Assignment No 1: Submitted To: Sir. Dr. Tair Zaidi.

The document implements the LMS and RLS adaptive filtering algorithms. It generates random input data, calculates the moving average to get the input signal, and sets the desired response to all ones. For LMS, it calculates the auto-correlation and cross-correlation matrices to get the weight vector over iterations. For RLS, it initializes the inverse correlation matrix, weight vector, and calculates the weight updates over iterations using the matrix inversion lemma. Plots of the desired, input, and error signals are shown.

Uploaded by

Syed Abdullah Hasan Chishti
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
67 views

Assignment No 1: Submitted To: Sir. Dr. Tair Zaidi.

The document implements the LMS and RLS adaptive filtering algorithms. It generates random input data, calculates the moving average to get the input signal, and sets the desired response to all ones. For LMS, it calculates the auto-correlation and cross-correlation matrices to get the weight vector over iterations. For RLS, it initializes the inverse correlation matrix, weight vector, and calculates the weight updates over iterations using the matrix inversion lemma. Plots of the desired, input, and error signals are shown.

Uploaded by

Syed Abdullah Hasan Chishti
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Mohammad

Arsalan
Khan
Assignment No 1:

Submitted to:

Sir. Dr. Tair Zaidi..


%%Implement of LMS algorithm for Adaptive Filters.

xi= rand(1000,1);

%%Order of Filter
N=5; (For better results, Higher order is considered)

for n=1:N
z(n) = 0;
for j=0:N-1
z(n) = z(n) + xi (n+j);
end
mov_avg(n) = z(n)/N;
end

%%Inputs
x=mov_avg

%%Desired Response
d=ones(1,N)

%%Auto Correlation Matrix of order N*N using Instanteneous Values


Rx=xcorr(x);

%%Cross Correaltion Matrix of order N*1 using Instanteneous Values


Ry=xcorr(x,d);

R = [Rx(1) Rx(2) Rx(3) Rx(4) Rx(5);


Rx(2) Rx(1) Rx(2) Rx(3) Rx(4);
Rx(3) Rx(2) Rx(1) Rx(2) Rx(3);
Rx(4) Rx(3) Rx(2) Rx(1) Rx(2);
Rx(5) Rx(4) Rx(3) Rx(2) Rx(1)]

P = [Ry(1);
Ry(2);
Ry(3);
Ry(4);
Ry(5)]

%%Scaling Factor
mew=[0.023; 0.055; 0.069; 0.091];

%%Weight Vector
w=zeros(1,5);

for n=1:N-1

%%Error Equation
e(n) = d(n) - w(n) * x(n);

%%Tap Update Equation


w(n+1) = w(n) + mew(n) * e(n) * x(n);
end
RESULTS:

x= 0.6293 0.7359 0.7871 0.7397 0.6863

d= 1 1 1 1 1

R= 0.4319 0.9705 1.5798 2.1321 2.5752

0.9705 0.4319 0.9705 1.5798 2.1321

1.5798 0.9705 0.4319 0.9705 1.5798

2.1321 1.5798 0.9705 0.4319 0.9705

2.5752 2.1321 1.5798 0.9705 0.4319

p= 0.6293

1.3652

2.1522

2.8919

3.5782
Implement of RLS algorithm for Adaptive Filters:-
rnd= rand(1000,1);
N=4; (Order of Filter)
t=[0:N-1];
for
i=1:N
x(i) = 0;
for
j=0:N-1
x(i) = x(i) + rnd (i+j);
end
mov_avg(i) = x(i)/N;
end

y=mov_avg (Inputs)
d=ones(1,N) (Desired Response)
ita=10^4;
I=ones(1,N);
R=ita*I;
w=zeros(1,N);
for
i=1:N
e(i) = d(i) - y(i);
z(i) = R(i) * x(i);
q = x(i)' * z(i);
v = 1/(1+q);
zz(i) = v * z(i);
w(i+1) = w(i) + e(i)*zz(i);
R(i+1) = R(i) - zz(i)*z(i);
end

subplot(3,1,1)
plot(t,d)
title('Desired Signal')

subplot(3,1,2)
plot(t,x)
title('Noise Corrupted Signal')

subplot(3,1,3)
plot(t,e)
title('Error')

y = [0.6802 0.7714 0.7502 0.7567]


d = [1 1 1 1]

You might also like