I am trying to plot magnitude against frequency averaged over a time window.

To do this I make a morlet wavelet Filterbank. And multiply the dft of each kernel by the dft of my incoming data set.

I take the mean value for each kernel multiplication result and plot this.

However somewhere along the way I have lost how to convert this to a meaningful scale.

I am using apple accelerate to perform the FFT, and as this does not scale, I divide each result by the sample size.

When I extract the magnitude using mag = sqrt((real^2) + (imag^2)) I am left with an array of very small numbers.

How do I convert this to meaningful magnitude in dB?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.