Clear Filters
Clear Filters

FFT and sin wave

4 views (last 30 days)
i Venky
i Venky on 11 Oct 2011
A small doubt in signal processing. When you take fft of a signal -- k=fft(y,512); and plot it, we get 512 points. I want to represent the 'x' axis in Hz . How would you do that? For eg: if k=40 what is the the corresponding frequency in Hz for a 512 point FFT?

Accepted Answer

Wayne King
Wayne King on 11 Oct 2011
If the signal is real-valued:
t = 0:.001:1-0.001;
Fs = 1e3;
x = cos(2*pi*100*t)+randn(size(t));
xdft = fft(x);
xdft = xdft(1:length(x)/2+1);
freq = 0:Fs/length(x):Fs/2;
plot(freq,abs(xdft));
xlabel('Hz');
  1 Comment
Wayne King
Wayne King on 11 Oct 2011
Frequencies are spaced at Fs/N where Fs is the sampling frequency and N is the length input signal.

Sign in to comment.

More Answers (2)

i Venky
i Venky on 11 Oct 2011
This is okay when the no of points in the fft is same as that of the length of the input signal. When I change the no of points in the fft. For eg: in your code if I change xdft=fft(x) into xdft=fft(x,512), what should be the change that I should make to get the frequency in Hz?

i Venky
i Venky on 11 Oct 2011
I got it.
Frequency in Hz= k * Sampling frequency (in time domain)/ N
Where 'N' is the no of points in the FFT.
  1 Comment
Wayne King
Wayne King on 11 Oct 2011
Yes, but just keep in mind that the frequency resolution of your DFT is determined by the length of your input signal, NOT by the value of any zero padding. Zero padding just interpolates your DFT, it does not give you any better resolution.

Sign in to comment.

Categories

Find more on Fourier Analysis and Filtering in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!