From: Ken B on
New Matlab user here. I have been given some data and asked to do a couple of things. First, was to take a FFT to find the frequency of an oscillation that should be present. The data is amplitude data as a laser was swept from frequency A to frequency B with step size N. the amplitude was recorded at every N. So I have a 1xN vector.

For laser jocks out there, I know that we should be able to see a frequency component corresponding to the FSR (free spectral range)

When looking in the Matlab help under FFT, it says I need to know the fs, or samples per unit time. How do I find/calculate this number?
From: Royi Avital on
"Ken B" <fiziksgeek(a)gmail.com> wrote in message <hs1aqc$k5d$1(a)fred.mathworks.com>...
> New Matlab user here. I have been given some data and asked to do a couple of things. First, was to take a FFT to find the frequency of an oscillation that should be present. The data is amplitude data as a laser was swept from frequency A to frequency B with step size N. the amplitude was recorded at every N. So I have a 1xN vector.
>
> For laser jocks out there, I know that we should be able to see a frequency component corresponding to the FSR (free spectral range)
>
> When looking in the Matlab help under FFT, it says I need to know the fs, or samples per unit time. How do I find/calculate this number?

Are you asking for the frequency each bin in the FFT vector represents?
If you have N samples sampled at Fs each bin is [num of Bin] * (Fs / N).
From: Ken B on
"Royi Avital" <RoyiREMOVEAvital(a)yahoo.com> wrote in message

> Are you asking for the frequency each bin in the FFT vector represents?
> If you have N samples sampled at Fs each bin is [num of Bin] * (Fs / N).

Not exactly, I think. I have a plot of amplitude vs. frequency with ~6000 points with a known spacing in frequency. Its sinusoidal in shape with some noise of course. I want to do a Fourier transform on it because the frequency of this sinusoidal wave should correspond to the free spectral range of a cavity in the system.

All the examples I have seen, use data that is a time vs amplitude signal. I have a time vs. frequency, but want to know the frequency of the frequency signal...if that makes sense...?
From: Godzilla on
"Ken B" <fiziksgeek(a)gmail.com> wrote in message <hs3t9p$n6n$1(a)fred.mathworks.com>...
> "Royi Avital" <RoyiREMOVEAvital(a)yahoo.com> wrote in message
>
> > Are you asking for the frequency each bin in the FFT vector represents?
> > If you have N samples sampled at Fs each bin is [num of Bin] * (Fs / N).
>
> Not exactly, I think. I have a plot of amplitude vs. frequency with ~6000 points with a known spacing in frequency. Its sinusoidal in shape with some noise of course. I want to do a Fourier transform on it because the frequency of this sinusoidal wave should correspond to the free spectral range of a cavity in the system.
>
> All the examples I have seen, use data that is a time vs amplitude signal. I have a time vs. frequency, but want to know the frequency of the frequency signal...if that makes sense...?

Why use an FFT?

I would just fit a sinusoid to the data.
From: Ken B on
"Godzilla " <godzilla(a)tokyo.edu> wrote in message <hs4954$iob$1(a)fred.mathworks.com>...
> "Ken B" <fiziksgeek(a)gmail.com> wrote in message <hs3t9p$n6n$1(a)fred.mathworks.com>...
> > "Royi Avital" <RoyiREMOVEAvital(a)yahoo.com> wrote in message
> >
> > > Are you asking for the frequency each bin in the FFT vector represents?
> > > If you have N samples sampled at Fs each bin is [num of Bin] * (Fs / N).
> >
> > Not exactly, I think. I have a plot of amplitude vs. frequency with ~6000 points with a known spacing in frequency. Its sinusoidal in shape with some noise of course. I want to do a Fourier transform on it because the frequency of this sinusoidal wave should correspond to the free spectral range of a cavity in the system.
> >
> > All the examples I have seen, use data that is a time vs amplitude signal. I have a time vs. frequency, but want to know the frequency of the frequency signal...if that makes sense...?
>
> Why use an FFT?
>
> I would just fit a sinusoid to the data.


That would work for the main frequency dependency. There is also some higher frequency modulation visible on top of the main sinusoidal signal, I think an FFT would give me all the info at once.

Plus is really bugs me that I cannot figure out how to do the FFT...haha