Faudy
Newbie level 3
periodogram distribution
I want to simulate a periodogram of a smoothed and averaged periodogram.
For that, I need the probability distribution of a such periodogram.
The averaged periodogram is a random variable of a Gamma distribution with parameters N (the number of averaged periodograms) and S(f)/N (S(f) is the PSD of the signal).
The smoothed periodogram consists in multiplying the signal by a window in the time domain. In frequency domain, this leeds to a convolution of the Fourier transforms.
Then, we must take the absolute value to have the periodogram estimation.
Any idea about it?
![Crying :cry: :cry:](data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7)
I want to simulate a periodogram of a smoothed and averaged periodogram.
For that, I need the probability distribution of a such periodogram.
The averaged periodogram is a random variable of a Gamma distribution with parameters N (the number of averaged periodograms) and S(f)/N (S(f) is the PSD of the signal).
The smoothed periodogram consists in multiplying the signal by a window in the time domain. In frequency domain, this leeds to a convolution of the Fourier transforms.
Then, we must take the absolute value to have the periodogram estimation.
Any idea about it?