clear;
bits = 64;
threshold = 0.5;

%Generate 64 random values. Check help on rand() function.
data = rand(1, bits);

%Quantize the random values only between two levels; 0 and 1.
for i=1:length(data)
    if data (i) > threshold
        data2(i) = 1;
    else
        data2(i) = 0;
    end
end

%Commands for plotting the data.
subplot(2, 1, 1); 
stairs(data2)
title('Original Digital Data');
ylabel('Amplitute');
axis([0 bits -0.2 1.2]);
grid off;


%%%%%%%%%%%%%%%Genrate Gaussian Noise%%%%%%%%%%%%%%%%%%%%%%
%%Different values for standard deviation for noise
sn = 1.2;

%%Generare different noise vectors.
noise1 = randn(1, bits)*sn;   
    
%%Add noise to original data;
recv_x = data2+noise1;
    
%%Quatize the received data back to 0's and 1's%%
for i=1:length(recv_x)
    if recv_x(i) > threshold
           recv_data(i) = 1;
    else    
        recv_data(i) = 0;
    end
end

%Note how the received data is different than the data sent.
subplot(2, 1, 2); 
stairs(recv_data)
title('Received Quantized Data');
ylabel('Amplitute');
axis([0 bits -0.2 1.2]);
grid off;

%Calculate the number of errors
error = XOR(data2, recv_data);
error_cnt = 0;
for i=1:length(error)
    if (error(i) == 1)
        error_cnt = error_cnt+1;
    end
end

%Calculate the error probability%%
error_prob = error_cnt/length(data2);

%Calculate the Signal-to-Noise Ratio%%
signal_strength = std(data2);
noise_strength = std(noise1);
SNR = 10*log10(signal_strength./noise_strength);

%Find tuples of (error_prob, SNR) for different values 
%of noise and plot them for every encoding scheme.