PDF, PSD, Auto-covariance, Fourier spectrum - all from .wav file
Asked by Tom on 24 Jan 2012
Latest activity Commented on by Matt Fig on 2 Nov 2012
Hi, I need to calculate the above for a given .wav file. I know how to arrive at each solution using mathematical techniques, but I also need to know how to arrive at each one using Matlab's own built in functions. My problem is there seem to be lots of different types of FFT functions and I can't find a clear reference to the others. Can anyone point me in the right direction as I'm pretty lost.
Products
1 Answer
Answer by Wayne King on 24 Jan 2012
Accepted answer
Hi Tom, you can use spectrum.periodogram to obtain a PSD estimate. You can use xcorr() to obtain an estimate of the autocorrelation sequence. You can use fft() to obtain the discrete Fourier transform.
Not sure why you want the PDF, The PSD is a form of a PDF for a wide-sense stationary random process, but it is not normalized. You could of course normalize it like a PDF, but I don't see any benefit in that.
0 Comments
1 Comment
Direct link to this comment:
http://www.mathworks.nl/matlabcentral/answers/26905#comment_108882
Hi, I need to calculate the above for a given .wav file. I know how to arrive at each solution using mathematical techniques, but I also need to know how to arrive at each one using Matlab's own built in functions. My problem is there seem to be lots of different types of FFT functions and I can't find a clear reference to the others. Can anyone point me in the right direction as I'm pretty lost.