How to get spectrogram of an array in MATLAB?

So.. I want to make spectrogram out of 4 signals that i need to add in array.

When I add them in array, it looks like this s=[x1 x2 x3 x4]; or with values: [1 1 0 1] But I'm not sure if it's proper adding..

Example of signal si this:

f1=1000; t1=0.2; x1=cos(2*pi*f1*t1);

so x1 is 1.

Is there a way of making spectrogram out of s?

Answer 1

[Star Strider]

The documentation states that the input signal is ‘specified as a row or column vector’, so you would need to loop through each column of ‘s’, saving the results (outputs you will need to request in your code) in appropriate cell arrays.

Example —

for k = 1:size(s,2)
    [sp{k},w{k},t{k}] = spectrogram(s(:,k));
end

Add other arguments to the spectrogram call as necessary to produce the results you want.

[Bridgit Schneider]

I get what you wanted to say, but I have formula from which I need to form 4 signals, add them in array and out of that array I need to make spectrogram.

What i made was:

f1=1000;
t1=0.2;
x1=cos(2*pi*f1*t1);
f2=2000;
t2=0.3;
x2=cos(2*pi*f2*t2);
f3=5/4;
t3=0.2;
x3=round(cos(2*pi*f3*t3));
f4=1000;
t4=0.2;
x4=cos(2*pi*f4*t4);
s=[x1 x2 x3 x4];

So I'm not sure how do I make spectrogram with that.. Can it be done with x=spectrogram(s,4); or something like that? I'm not sure which command do I use..

[Star Strider]

I was under the impression that ‘t1’ and the others were equal-length vectors, and the elements of ‘s’ were column vectors. Instead, ‘s’ is a (1x4) vector. It is difficult (almost impossible) to even do a Fourier transform on onlly 4 points.

It is not possible to do a spectrogram on ‘s’ as it currentlly exists. You need to re-define it to be something meaningful.

[Bridgit Schneider]

Sorry, I didn't know how to describe question and i could've just post the code directly like that but okay..

Is there any way of using those signals in other way and making spectroram out of them?

[Star Strider]

It depends on what the ‘other way’ is. They need to be vectors.

Here’s an example:

Fs = 8192;                                      % Sampling Frequency (Hz)
ft = @(d,sf) linspace(0, d*sf, d*sf)/sf;        % Time Vector Function
f1=1000;
t1=0.2;
x1=cos(2*pi*f1*ft(t1,Fs));
f2=2000;
t2=0.3;
x2=cos(2*pi*f2*ft(t2,Fs));
f3=5/4;
t3=0.2;
x3=round(cos(2*pi*f3*ft(t3,Fs)));
f4=1000;
t4=0.2;
x4=cos(2*pi*f4*ft(t4,Fs));
s=[x1 x2 x3 x4];

figure
spectrogram(s,[],[],[],Fs)

The ‘ft’ function creates time vectors given time duration ‘d’ and sampling frequency ‘sf’. In then calculates these for each segment of the ‘x’ values that eventually create ‘s’.

I’m not certain what you’re doing or what you want as the result. This example should get you started.

[Bridgit Schneider]

I think this will help a lot! Thank you very much for help :)

[Star Strider]

As always, my pleasure!