It's all about the Software

I have a channel sounder consisting of 2 USRPs X410, each connected to its corresponding host PC. Both USRPs are connected to an Octoclock, for frequency and time synchronization. Nevertheless, for better time synchronization, there are additional time commands (https://uhd.readthedocs.io/en/latest/pa ... dcmds.html, https://kb.ettus.com/Synchronizing_USRP ... nds_in_UHD), which in theory should synchronize both devices in the next rising edge of the pps, and the time the tx and rx start transmitting/receiving. Despite the time commands, I see that whenever I repeat the measurement, the channel impulse response and the phase of the channel frequency response change. The measurements were conducted using a cable. I would have expected to be able to repeat the same measurement. Am I maybe overlooking something? I am modifying the uhd examples to transmit and receive. Thank you for your help

This is not the best place to be asking this. This is for Simon's SDR-Radio suite of SDR software tools. Your problem is WAY out of the usual for this forum. I'd suggest that you arrange for your USRPs to share a single clock oscillator if you need to synchronize them to each other precisely. Otherwise you're pissing up wind trying to get any meaningful synchronization between the radios. I am picky - phase locking the two oscillators to GPS very probably won't do the trick without some heartburn tuning out the offsets in the PLLs. (PLLs lock frequencies at a slight phase offset. The phase offset is what locks the usual second order loops. A third order loop might be suitable but has its own problems.)

{^_^} (No, Joanne's not likely to dig into this one further due to lack of equipment to work on it properly even though the concepts here are part of my home turf.)