Hey Dennis, so UHD is reporting valid calibration files -- that's a good start. How are you generating a signal with a specific power level? Are you doing `uhd_siggen -p -10` or something like that?
Also, how did you install UHD? If you installed from source (which we don't generally recommend if you can install from binaries, so I'm not expecting this) then you can crank up the log level and watch the power API operate in real time (export UHD_LOG_CONSOLE_LEVEL=trace, assuming you built with `cmake -DUHD_LOG_MIN_LEVEL=trace`). --M On Wed, Sep 17, 2025 at 10:29 AM <[email protected]> wrote: > Hi Martin, > > I am a bit stuck here (still). I performed several measurements using an > OFDM signal (20 MHz, synthetic signal) and a sine wave on a smaller > bandwidth (1 MHz, via uhd_siggen) to verify the effectiveness of the B210 > RX calibration files that I had created. All in a controlled, cabled setup > with a 30 dB attenuator in place. To this point, it is hard to tell, as I > see little to no difference in power when loading the calibration files. > Therefore, I have three hypotheses: > > 1) These values are so small that the power calibration is negligible. > > 2) The calibration file is not loaded correctly for the B210. Missing > something in the code (I only get a True or False when found or not found) > > 3) A specific value in the calibration file is not loaded because of > frequency offsets, which, by default, exist (I measured this), and so the > value of the calibration file is not used. > > > Is there another way to determine if a specific power value from the > calibration file is loaded? Rather than getting the message now: "I am a > B210 and found a valid calibration file for channel X or Y". > > Could you give me some advice on these points? Or am I missing some other > critical point here? > > Thank you! > _______________________________________________ > USRP-users mailing list -- [email protected] > To unsubscribe send an email to [email protected] >
_______________________________________________ USRP-users mailing list -- [email protected] To unsubscribe send an email to [email protected]
