Hi,

This answer made me curious. Is it because it is not implemented yet or because you need an exact (same) clock for the Rx and Tx part. Or is it because of something else.

You need the exact same clock (or at the very least phase locked ones, you could have different sample rates if they are locked you can always resample in software). Then you also need to know absolute time sync (i.e. which sample of TX matches exactly which sample on the RX side).

Theses requirements, while not impossible to achieve would impose some hardware modifications to whatever cheap SDR you try to use (like hackrf for TX and RTL-SDR for RX), making it hard to reproduce. It would also required _a_lot_ of work on the software side to deal with all the peculiarities of such a system. And finally there are cheap-ish full duplex SDR.

So simply put, there is too high a cost and too low a reward to bother. I mean, if someone has a spare 100kEUR in their couch cushion to finance the dev, I'm sure I can get that implemented.

Cheers,

Sylvain