How to get amplitude-frequency response curve?

How should one go about plotting an amplitude-frequency curve?

So far, what I've done is to take a signal, take the upper and lower envelope peaks (using envelope() function) and subtract them to get my amplitude. I would take the fft of the original signal in order to get my frequency. I also made sure that the fft I took was of the same vector length as my amplitude. However, whenever I plot the two I get a blank graph. Why is this?

[Image Analyst]

You say you "take the fft of the original signal" and then you "plot the two". Two what? Real and imaginary parts? What does the envelope have to do with any of that?

And I don't understand why you subtract the envelope from the original signal. What does that get you? You already know the amplitude at any point because you have the original signal.

[Chinwe Orie]

The reason why I take the envelope peaks is because the amplitude of this signal is not 6.155. It is half the difference between 6.155 and 6.125 (approximately).

I got better numbers, however when I just substracted the mean of the signal from the signal itself and presented that as my amplitude. I think this may be the better way, but I was instructed to do it the first way.

By the two, I mean the amplitude and frequency (the real part of the fft, not the imaginary part. This is the code. It's from a function I created with some code I found on MATLAB help. This gave me a pretty decent graph, but it was when I used the envelope peaks that it gave me the blank graph.

[d,~,~] = xlsread(x);
t = d(:,1);
s = d(:,2);
fin = isfinite(s);
t = t(fin);
s = s(fin);
% Ts = mean(diff(t));
Fs = rate;
L = size(t,1);
subplot(4,1,1)
plot(t, s)
grid
title('Original Signal')
sm = s-mean(s);
FTs = fft(sm)/L;
Fv = linspace(0, 1, L)*Fs;
Iv = 1:numel(Fv);
y = abs(FTs(Iv))*2;
subplot(4,1,2)
plot(Fv,y)
grid
xlim([0  300])
title('Fourier Transform of Baseline-Corrected Signal')

[Image Analyst]

Well I'm glad you're not using the envelope function like you said. You're merely subtracting the mean. Which is fine, but not necessary. You could leave it in there, you'd just have a huge spike at the zero frequency location.

I can't really try your data because it seems you removed your workbook. All I can say is to look at the y values and try to see what their values are. Is one value huge, like the first or last one and so, because of that, you can't see any in-between values?

Attach your workbook again if you want anyone to help you.

[Chinwe Orie]

Here it is. The rate, or Fs is 25000.