Hi everyone, I am trying to implement an 8-input correlator on a ZCU208. The CASPER correlator tutorials and some of you have helped me a lot, and I am currently testing with a couple of inputs. ADC sampling at 3.25 Gsps 2^14 points 8-taps PFB and FFT_wideband_real.
[image: Screenshot 2025-04-21 at 10.31.32 AM.png] The output spectrum shows a frequency shift, for eg. Here's a 450MHz tone; it should have been a straight line. I was told that it could be due to a mistake in the reset logic surrounding new_acc and vector accumulation. I have 4 parallel streams of FFT outputs (each 2k), so my vector accumulation length is 2048, and I interleave 4 output registers to construct the 8k spectrum. If I reduce the accumulation to 512 samples, then it works fine and the tone doesn't shift in frequency. Maybe I am missing something basic here, does it have to do with delaying the signal before accumulating? Has anyone encountered this in the past? thanks --Liju -- 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 view this discussion visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CA%2BrQCpZAyFD9OGm7QMSndrB0vW2CzoBeOwZRqaVbVFmaKHVWGA%40mail.gmail.com.
