First, let me start by saying, I feel very silly asking this question but here it goes:
Say I have a signal post ADC, and I split that signal and baseband (baseband different copies at different Nyquist zones using a complex phasor, nothing fancy) multiple copies of that signal and combine it, would I get a SNR improvement? I sense that the answer is NO as the noise is correlated (in fact perfectly correlated) so there is no benefit.
I wanted to see if anyone thinks differently.
Thanks in advance!
That is not silly question and it reminds me of concept of diversities. I once asked how come we can get multiple/infinite number of copies of a signal without paying for energy rule. I didn't get answer but then imagined two mirrors reflecting for ever...it is light travelling until it vanishes and our interpretation is images, still the analogy doesn't make sense to me.
I don't know what you mean by "baseband different copies at different Nyquist zones using a complex phasor", or "multiple copies of that signal".
Until you explain what you mean there it might be tough to answer your question.
I put this graphic together in the hope to clarify the question. Basically, summing A and B after bringing the aliases to DC
Remember that A and B are mirror images of each other, so just mixing them down separately and adding them would be a bit of a mess.
Even if you reverse one before adding them together, even if you could add them completely coherently you won't be adding any information, so there would be no increase in SNR or reduction in any impairments. You'd just be doubling the amplitude.
Thank you very much for your reply. Appreciate it!