Redpitaya Forum

I opened this issue on gr-osmosdr the other day: https://gitea.osmocom.org/sdr/gr-osmosdr/issues/3

I noticed that in gr-osmosdr there is a check to see that the center frequency is greater than samp_rate/2. I can't really see why this is a necessary sanity check. Additionally, the server side code, sdr-transceiver-wide.c for instance, has a minimum frequency of 0Hz.

I'm still learning about SDR, but isn't that so it is within the Nyquist frequency?

I'm guessing the minimum of 0Hz is because the Red Pitaya can sample down to DC if the STEMLab boards are used or down to 300 kHz for the SDRLab.

John M0JPI

Hey John

Any signal of interest, after mixing down to baseband, needs to be below samp_rate/2 to avoid aliasing.

However, in this case, I don't think we should be worried about sampling yet as this frequency (the center frequency) is simply the frequency that will be produced by the DDS, and therefore is the frequency in the received/transmitted signal that will be shifted down to 0Hz.

Below is the diagram for the wideband SDR notes page. The frequency mixing occurs at the full sample rate of 125MSPS. After the complex multiply the center frequency has been shifted to 0Hz. At this point its still 125MSPS. Only after this shift is downsampling and lowpass filtering performed, at which point your band of interest will need to be below samp_rate/2.

I suppose, regardless of whether my assessment is right or wrong, there is a discrepancy between the board side code and gr-osmosdr.



-Jay