I have a question about power spectral densities of time signals and negative frequencies. Lets say I generate some white noise with an arbitrary waveform generator. If I would study its power spectral density I would just look at the positive frequency part; for a real signal the negative frequencies are redundant. But in nature one encounters systems where this is not true; many quantum mechanical systems have a power spectral density that is not symmetric around 0 frequency.

Is there a way I could use an AWG to generate some voltage signal that has such an asymmetric power spectral density? Perhaps using two channels and a phase difference? If so, how would you characterize this? Not using a spectrum analyzer I assume

I'm asking because this would be extremely interesting in physics experiments where one wants to simulate a quantum mechanical environment, which is not trivial to engineer