psd normalization


Uri
07-18-2007, 09:39 AM
I'm having problems with the psd resulting from spctrm.cpp function descrived in chapter 13.

I'd like to compare function's result with a matlab computed psd and, althougth psd trend is the same, it seems to be a multiplying factor there.

Of course I've tryed to divede both psd to obtain that factor, but it seems not that easy so I find out is not a constant at all.

Anyone may know what's happening?

Thanx

I attach figures obtained for a better undertanding.

juancmaya
04-30-2011, 11:21 PM
i have the same results for low frecuency , when i work with a sound signal sampled to 40khz i get only a peak when the p[j] graphic start, i have m=128 k=100 ovrloap=0, what do you think?

mayajuan2002@hotmail.com


I'm having problems with the psd resulting from spctrm.cpp function descrived in chapter 13.

I'd like to compare function's result with a matlab computed psd and, althougth psd trend is the same, it seems to be a multiplying factor there.

Of course I've tryed to divede both psd to obtain that factor, but it seems not that easy so I find out is not a constant at all.

Anyone may know what's happening?

Thanx

I attach figures obtained for a better undertanding.