Hi Alex, I don't know if there's a way to avoid generating the full spectra prior to taking the fft (I suspect if there is, it just pushes the buffering somewhere else), but it certainly seems like you should be able to do FFTs of two streams, A, B, in parallel. I would think this would work by (after generating the full spectra) computing FFT(A+iB) such that FFT(A) emerges in the real part of the output, and FFT(B) in the imag. If you don't need to FFT two independent streams you can probably do something smart (like the wideband real fft) to leverage the above to do the multiple serial FFTs on a wideland input.
I think? I haven't had coffee yet. Jack On Mon, 13 Aug 2018, 11:36 pm Alexander Raymond, < alexander.raym...@cfa.harvard.edu> wrote: > Dear CASPER List, > > I am a postdoc working with Jonathan Weintroub and am new to CASPER. > > We are working on bitcode for ROACH2 to reformat the data product at the > Submillimeter Array in real time so that it can be used directly in VLBI > observations as part of the Event Horizon Telescope. As part of that > reformatting, we wish to convert the demultiplexed frequency domain streams > of the SWARM F-engine (https://arxiv.org/abs/1611.02596) to time domain > streams. In other words, we want to take the inverse FFT of demultiplexed > frequency-domain samples. > > Ordinarily, we would would achieve the inverse fft using the CASPER > complex_conj and fft blocks in series; however, the SWARM frequency domain > output comes from an fft_wideband_real block, which discards the negative > frequencies, so direct application of complex_conj+fft will not work. One > approach to inverting the data would be to buffer and copy the positive > (complex) frequencies emitted by fft_wideband_real, assemble a stream of > positive and (mirrored) negative frequencies, then feed the complete > spectrum into complex_conj+fft blocks. > > Jonathan suggested I check with the CASPER community to see, is there a > more efficient way of performing our inverse FFT? Is there a way to take > advantage of the knowledge that the output from our iFFT block should be > real-valued? > > We appreciate any guidance the mailing list can provide. > > Sincerely, > > Alex Raymond > > -- > You received this message because you are subscribed to the Google Groups " > casper@lists.berkeley.edu" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to casper+unsubscr...@lists.berkeley.edu. > To post to this group, send email to casper@lists.berkeley.edu. > -- You received this message because you are subscribed to the Google Groups "casper@lists.berkeley.edu" group. To unsubscribe from this group and stop receiving emails from it, send an email to casper+unsubscr...@lists.berkeley.edu. To post to this group, send email to casper@lists.berkeley.edu.
