How do i get the Nyquist frequency from FFT and PSD?

Here is the code I was provided. I am unsure how to get the Nyquist frequency from the plot.

h_11 = 0.4*sinc(0.4*[-5:5]);

Npt=512;

fft_h_11 = fft(h_11,Npt); %FFT of filter

psd_h_11 = fft_h_11 .* conj(fft_h_11); %PSD of filter

psd_h_11_dB = 10*log10(psd_h_11); %PSD in dB

h_41 = 0.4*sinc(0.4*[-20:20]); %Impulse response, N=41

fft_h_41 = fft(h_41,Npt); %FFT of filter

psd_h_41 = fft_h_41 .* conj(fft_h_41); %PSD of filter

psd_h_41_dB = 10*log10(psd_h_41); %PSD in dB

omega = [1:Npt]*20/Npt; %frequency axis

subplot(2,1,1) %create an array of plots, 2 by 1

plot(omega,psd_h_11,omega,psd_h_41) %plot linear-scaled PSD

legend('10th order', '40th order')

title('Example 9-1 Linear Scale')

subplot(2,1,2) %Activate lower plot

plot(omega,psd_h_11_dB,omega,psd_h_41_dB) %plot logarithmic-scaled PSD

legend('10th order', '40th order')

title('Example 9-1 Logarithmic Scale')

Answer 1

[Star Strider]

You do not get the Nyquist frequency from the plot, you calculate it from the data.

The Nyquist frequency is half the sampling frequency.

So:

t = ...;                  % Time Vector
Ts = t(2)-t(2);           % Sampling Interval (Assuming Uniform Sampling)
Fs = 1/Ts;                % Sampling Frequency
Fn = Fs/2;                % Nyquist Frequency

I leave the rest to you.

[Jennifer Canady]

Thank you

[Star Strider]

As always, my pleasure.

If my Answer helped you solve your problem, please Accept it!

[Cédric Cannard]

Is there a way to estimate it without sampling frequency and time vector? Wouldn't the power spectrum fall down at the Nyquist frequency?

[Star Strider]

If you know that the highest displayed frequency is the Nyquist frequency, you can infer the sampling frequency and sampling interval from it.