# Plotting FFT of several data sets (signals) of same length, in the same graph

17 views (last 30 days)
Muhammad Shafiq on 12 Sep 2018
Commented: Muhammad Shafiq on 16 Sep 2018
I have a data of several thousands samples (say 60000), the sampling frequency is given (1 MHz). I want to divide the data in equal frames, each of 500 samples that makes 120 frames, then plot all those (120) frames on the same graph to compare them. I want to plot in time domain and frequency domain. I have been able to do the time domain plot but I don’t know how to do the frequency domain plot of all of the signals in the same graph. Looking forward for the help. Thanks

dpb on 12 Sep 2018
Might look at waterfall plot.
I don't know what you mean as far as "in the same graph" when mixing time and frequency domains??? Can you show a representative plot of what you'd like yours to look like?
Muhammad Shafiq on 12 Sep 2018
I mean time plots in one graph and frequency domain plots in another. For example;
plot(f/1E6,2*abs(Y1(1:NFFT/2))/1E-3,'r', ...
f/1E6,2*abs(Z1(1:NFFT/2))/1E-3,'b', ...
f/1E6,2*abs(A1(1:NFFT/2))/1E-3,'g');
This plots the FFTs of three signals in same graph. But if we have large number of signals then how to do it. While signals are, as I mentioned in my main question.

Aquatris on 12 Sep 2018
You put all of the signals you want in a matrix (each column is a different partition of the signal);
data = rand(60000,1); % replace with your data
y = [];
for i = 1:120
y(:,i) = data((i-1)*500+1:(i)*500);;
end
% perform the fft as;
L = 500;
Y = fft(y);
P2 = abs(Y/L);
P1 = P2(1:L/2+1,:);
P1(2:end-1,:) = 2*P1(2:end-1,:);
% obtain frequency matrix;
Fs = 1e3; % sampling frequency
f = Fs*(0:(L/2))/L;
% plot frequency domain
figure(1)
semilogy(f,P1)
% plot time domain
figure(2)
plot([1:L]/Fs,y)
Make sure you understand what is done in each step and if you have questions, feel free to ask.

#### 1 Comment

Muhammad Shafiq on 16 Sep 2018
Many thanks Aquatris for your help and sorry for the late feed back I was on travel so responding late.
I have included my data file
Data1=VarName1(1:60000);% I used 60,000 samples among the big data file.
Then in your code I replaced my data
data = Data1; % replace with your data
y = [];
for i = 1:12
y(:,i) = data((i-1)*500+1:(i)*500);;
end
-------------------
I have not replaced anything in
y = [];
Still I could see the plots in time and frequency domain. Could you please mention further about it (y = [];)

Aquatris on 16 Sep 2018
y = [] just makes y an empty matrix. It is not a necessary code. I include it in the off chance that you had a variable named y in your code. That line would delete any contents of previously assigned y variable values

#### 1 Comment

Muhammad Shafiq on 16 Sep 2018
OK great and thanks again :)