Model an FPGA FM Radio Demodulator in Python

[Jyan]

I am trying to model an FPGA FM radio demodulator in Python. I will have a mixer that converts an FM channel to 36MHz IF, and I plan on undersampling this IF signal. If in the digital world (my Python model), I have a high sample rate FM signal at 1.2MHz, should I be able to resample to a lower sample rate (after using a BPF ofc) and obtain the same aliasing down to baseband?

My understanding is that this should be possible, it just doesn't seem to be working. I am confident I understand undersampling 100%, but am not so sure about resampling... When I resample the FM modulated signal I get complete garbage at the baseband. My high sample rate is 3MSPs, the FM signal is sitting with 200KHz bandwidth and a centre at 1.2MHz. I am resampling to 525KSPs, if this were undersampling then the FM signal would be sitting in the 5th nyquist zone. But does it work the same digitally?

 

[Greg Bernhardt]

Thanks for the post! Sorry you aren't generating responses at the moment. Do you have any further information, come to any new conclusions or is it possible to reword the post?

 

[Jyan]

I can add some more details. For now I have kind of sidestepped the problem.

The essential question is whether the effects of undersampling, (see: https://en.wikipedia.org/wiki/Undersampling) can be replicated in the discrete time domain via resampling (see: https://en.wikipedia.org/wiki/Decimation_(signal_processing) , https://en.wikipedia.org/wiki/Upsampling). Currently, I think it is actually not possible, resampling and undersampling are fundamentally different, but I'm not certain. I am by no means an expert in digital signal processing.