Finding power spectral density using MATLAB,

[geft]

Question:
Using Matlab, generate random unary and bi-polar sequences of period Tb=1s, over-sample each sequence by a rate of 4, calculate and display their PSD.

It took me way too long to generate proper unary sequences. I have yet to figure out how to code bipolar sequences.

Here is the code I came up with. I don't understand half of it and the result doesn't look correct. The problem is that the psd appears completely flat at 0 amplitude at high frequencies.

h = commsrc.pattern('SamplingFrequency', 1, 'SamplesPerSymbol', 10, 'PulseType', 'NRZ', 'OutputLevels', [0 1], 'DataPattern', 'PRBS5');
x = generate(h, 1000);                    
plot(x)

Hpsd = dspdata.psd(x, 'SpectrumType', 'Twosided', 'CenterDC', true, 'Fs', 4);
plot(Hpsd)

 

[KataKoniK]

First of all, great job on generating the unary sequences! It can definitely be a bit tricky to figure out at first. As for the bipolar sequences, here is some code that should help you generate them:

% Generate bipolar sequence of length 1000
bipolar = 2*randi([0,1], 1, 1000)-1;
% Plot the sequence
plot(bipolar)

Now, to oversample the sequences by a rate of 4, you can simply use the "upsample" function in Matlab:

% Upsample unary sequence by a rate of 4
unary_upsampled = upsample(unary, 4);
% Upsample bipolar sequence by a rate of 4
bipolar_upsampled = upsample(bipolar, 4);

To calculate and display the PSD, you can use the "periodogram" function in Matlab:

% Calculate PSD of unary sequence
[unary_psd, f] = periodogram(unary_upsampled, hamming(length(unary_upsampled)), length(unary_upsampled)*4, 1);
% Plot PSD
plot(f, unary_psd)

% Calculate PSD of bipolar sequence
[bipolar_psd, f] = periodogram(bipolar_upsampled, hamming(length(bipolar_upsampled)), length(bipolar_upsampled)*4, 1);
% Plot PSD
plot(f, bipolar_psd)

This should give you the correct PSD plot for both the unary and bipolar sequences. Hope this helps!