Hello all,

I have a question about the pnoise simulation.

I understand that in frequency domain, the switching system should translate noise in the following way:

cross correlate the noise source spectra with the model of the system large signal behavior (eg. square or sin signal in a mixer)

However, I read in the manuel that the pnoise is actually "a time-average of the noise at the output is computed as a spectral density versus frequency". This I could not understand, because it looks to me that "the model of the system large signal behavior" as mentioned previously is different from "a time-average of the system transfer function at each bias point in a period".

For example, for a simple mixer, "the model of the system large signal behavior" could be a square wave, while at each time point, this simple model results in a transfer with infinite BW, thus the "time-averaged" version also having infinite BW.

I don't know where goes wrong. Could anybody help me? Thanks.

Hello Ken,

Thanks very much for your reply. Now it is clear for me:)

Best regards,

Chen