Code 1 w = 10;

subplot(2,1,2);
plot(w,xw)
xcw2(i)=
trapz(t,xct2.*exp(-
close all; sigma=var(x)

clc; t = -10:0.01:10; %axis([-5 5 -5 10]); j*w(i).*t)); N=100000;

clear all;
close all;
t0 = 3; title('Fourier end x=rand(1,N); Code 4
xct = rectpuls(t-t0,w); Transform'); clc;
xlabel('w'); subplot(2,2,4); step=0.1; close all;
l = 10; subplot(2,1,1); ylabel('X(w)'); plot(w,xcw2) Range=-3:step:3;
t = -10:0.01:10; clear all;
plot(t,xct) axis([-5 5 -5 10]);
axis([-10 10 -1 2]); title('Fourier h=histc(x,Range); N=10000;
xct = rectpuls(t,l); title('Rectangular Transform'); pdf=h./(step.*sum(h)); x=randn(1,N);
Pulse'); xlabel('w');
subplot(2,2,1);
plot(t,xct)
xlabel('t'); Code 2(c) ylabel('X(w)'); %figure(1) nu=mean(x);
ylabel('x(t)'); clc; %bar(Range,h,'histc'); sigma2=var(x);
axis([-6 6 -1 2]); %histc is used to make
title('Rectangular clear all;
w = -8*pi:0.01:8*pi; close all; a function that divides step=0.1;
Pulse'); for i=1:length(w) %Taking Conv data into bins of size
xlabel('t'); range=-3:step:3;
xcw(i)= l = 10; xct3 = conv(xct1,xct2, equal
ylabel('x(t)'); trapz(t,xct.*exp(- 'same')*0.01; %to step size and gives
t = -10:0.01:10; h=histc(x,range);
j*w(i).*t)); figure(2); numbr of x values in
w = -8*pi:0.01:8*pi; end subplot(2,2,1); that bin
%First square pulse f=((1./sqrt(2*pi*sigma2
xct1 = rectpuls(t,l); plot(t,xct3); ).*exp((-((range-
for i=1:length(w) subplot(2,1,2); title('Conv');
xcw(i)= subplot(2,2,1); nu).^2))./(2.*sigma2)))
plot(w,xcw) plot(t,xct1) xlabel('t'); figure(2) );
trapz(t,xct.*exp(- axis([-5 5 -5 10]); ylabel('x(t)'); subplot(2,1,1);
j*w(i).*t)); axis([-6 6 -1 2]);
title('Fourier title('Rectangular plot(Range,pdf,'*','mar subplot(2,1,1);
end Transform'); w = -8*pi:0.01:8*pi; kersize',10);
Pulse'); plot(range,f,'b','linew
xlabel('w'); xlabel('t'); for i=1:length(w) idth',2);
subplot(2,2,2); ylabel('X(w)'); xcw3(i)=
plot(w,xcw) ylabel('x(t)'); hold('on');
trapz(t,xct3.*exp(- b=1;
axis([-5 5 -5 10]); j*w(i).*t)); a=0;
title('Fourier w = -8*pi:0.01:8*pi; simulated_pdf=
for i=1:length(w) end h./(step.*sum(h));
Transform'); N=size(h);
xlabel('w'); xcw1(i)= plot
trapz(t,xct1.*exp(- subplot(2,2,2); (range,simulated_pdf,'*
ylabel('X(w)');
Code 2 (b) j*w(i).*t)); plot(w,xcw3) N=N(1,2); ','markersize',15);
subplot(2,2,3); end hold on;
clc; title('Fourier hold on;
plot(w,abs(xcw)) Transform'); syms X;
clear all; subplot(2,2,2); grid on;
axis([-5 5 0 10]); xlabel('w'); f=rectangularPulse(a,b,
close all; plot(w,xcw1)
title('Magnitude ylabel('X(w)'); X);
Spectrum'); axis([-5 5 -5 10]); %axis([-3,3,0
w = 10; title('Fourier fplot(f); max(simulated_pdf)])
xlabel('w');
t = -10:0.01:10; Transform'); axis([-3 3 0 0.5]);
ylabel('|X(w)|'); cdff=zeros(1,N);
t0 = exp(j*10*t); xlabel('w');
xt = rectpuls(t,w).*t0; ylabel('X(w)'); multi = xcw1.*xcw2; cdff(1)=h(1)/sum(h); title('gaussian
subplot(2,2,4); subplot(2,2,4);
plot(w,angle(xcw)) distribution')
subplot(2,1,1); plot(w,multi) for i=2:N xlabel('range');
axis([-6 6 -4 4]); cdff(i)=cdff(i-
plot(t,xt) %Second square pulse ylabel('pdf');
title('Phase title('Fourier 1)+h(i)/sum(h);
axis([-6 6 -1 2]); xct2 = rectpuls(t,l); legend('theoratical','s
Spectrum'); Transform'); end
title('Rectangular imulation');
xlabel('w'); xlabel('w');
Pulse'); subplot(2,2,3);
ylabel('angle(X(w))'); ylabel('X(w)'); subplot(2,1,2)
xlabel('t'); plot(t,xct2) subplot(2,1,2);
ylabel('x(t)'); axis([-6 6 -1 2]); plot(Range,cdff,'*','ma
title('Rectangular rkersize',15) N=size(h);
w = -8*pi:0.01:8*pi; Pulse'); axis([-3 3 0 1.2]) N=N(1,2);
for i=1:length(w) xlabel('t');
Code 2(a) xw(i)= ylabel('x(t)'); hold on syms t z;

clc;
trapz(t,xt.*exp(-
j*w(i).*t));
Code 3 cdf=int(f,X,-3,X);
fplot(cdf);
p=(1./(sqrt(2*pi)))*int
w = -8*pi:0.01:8*pi; clc; ((exp(-(t*t/2))),t,-
clear all; end for i=1:length(w) 3,z);
close all; clear all; u=mean(x) fplot(p,'linewidth',2);

hold ('on'); xq = round(x*(2^(q- Delta = 2./(L-1); a=rand(1,N)>0.5; % Calculate the number

1))/A)*(A/(2^(q-1))); % Step size of the of valid code words N
simulatedcdf(1)=(h(1)/( quantization levels s=2*a-1; and the code rate RC
sum(h))); % Compute the SQNR = N = 2^k;
for i=2:N quantization error and zeros(length(L),1); % snr_dB=1:1:10; RC = k/n
the SQNR To store the Signal to n=1/sqrt(2).*randn(1,N)
simulatedcdf(i)=simulat e = x - xq; Quantization Noise ; % Generate all possible
edcdf(i-1) + sqnr(q) = Ratio (SQNR) for each combinations of k
h(i)/(sum(h)); 10*log10(var(x)/var(e)) quantization level for i = information bits
end; ; 1:length(snr_dB) information_bits =
end % Non-Uniform dec2bin(0:N-1)-'0'
plot(range,simulatedcdf Quantization y=10^(snr_dB(i)/20).*s+
,'*','markersize',10) % Plot the SQNR versus for i=1:length(L) n; % Apply the generator
axis([-3 3 0 quantization levels q = zeros(size(x)); a_dec=real(y)>0; matrix G to each
max(simulatedcdf)]) plot(1:Q, sqnr, 'o-') % Compute combination to get the
title('cdf of gaussian xlabel('Quantization quantization levels nErr(i)=size(find(a- corresponding n code
distribution') Levels') V = [-(L(i)-1)/2 : a_dec),2); bits
xlabel ('range'); ylabel('Signal-to- 1 : (L(i)-1)/2] * end code_words =
ylabel('cdf'); Quantization-Noise Delta(i); simBer = nErr/N; mod(information_bits*G,
legend('theoretical','s Ratio (dB)') % Quantize the theoryBer = 2);
imulated') title('Uniform signal 0.5*erfc(sqrt(10.^(snr_
Quantization SQNR vs. for j=1:N dB/10))); % Display the complete
Quantization Levels') [val, index] = code set C
min(abs(x(j)-V)); figure disp('Complete code set
q(j) = semilogy (snr_dB, C:');
Code 5 V(index); theoryBer, 'b-', disp(code_words);
%Write a matlab program Code 6 end
% Compute the SQNR
'LineWidth', 1.5);
hold on
to compute the Signal % Program to Compute noise = x - q; semilogy (snr_dB,
to quantization Noise SNR of Non-Uniform
ratio of Uniform Quantization and Plot
signal_power =
sum(x.^2)/N;
simBer, 'x',
'MarkerSize', 8); Code 8(b)
Quantization. Plot SNQR the SNR vs. noise_power = grid on clear all;
versus Quantization Quantization Levels sum(noise.^2)/N; legend('theory', clc;
levels SQNR(i) = 'simulation'); close all;
clear all; 10*log10(signal_power/n xlabel('snr_db -->');
% Set the parameters close all; oise_power); ylabel('bit error rate- G = [1 1 0 0 1 0 1;
N = 1000; % number of clc; end ->'); 0 1 1 1 1 0 0;
samples title('bit error 1 1 1 0 0 1 1];
fs = 1000; % sampling % Signal Parameters % Plot the SNR vs. probability curve for
frequency (Hz) N = 10000; Quantization Levels bpsk modulation'); G (3, :) = mod (G (3,
f = 10; % signal % Number of samples in figure; :) +G (1, :) , 2) ;
frequency (Hz) the signal plot(b, SQNR, 'b-o', G (2, :) = mod (G (2,
A = 1; % signal f = 1; 'LineWidth', 2); :) +G (3, :), 2) ;
amplitude % Signal frequency xlabel('Number of G (1, :) = mod (G (1,
Q = 10; % maximum Fs = 1000; Bits');
Code 8(a) :) +G (2, :) , 2) ;
quantization level % Sampling frequency ylabel('Signal to clc;
t = (0:N-1)/Fs; Quantization Noise clear all; G
% Generate the signal % Time vector Ratio (dB)'); close all;
t = (0:N-1)'/fs; % time x = sin(2*pi*f*t); grid on;
vector % Signal % Define the generator
x = A*sin(2*pi*f*t); % matrix G
sinusoidal signal % Quantization G = [1 1 0 0 1 0 1;
Parameters 0 1 1 1 1 0 0;
% Compute the SQNR for L = 2:20; Code 7 1 1 1 0 0 1 1];
each quantization level % Number of clc;
sqnr = zeros(Q, 1); quantization levels to clear all; % Calculate the number
for q = 1:Q try close all; of information bits k
% Quantize the b = log2(L); and code bits n
signal % Number of bits to N = 10^6; [k, n] = size(G);
represent each level