How to use bladeRF-cli's calibration function?

[ehhewng07]

Hi,
After read the help information of bladerf-cli calibration routines, found that still not work. I'm now running srsue and srsenb, found that the project lack this part code. Anyone can give me some examples on how to do DC and IQ imbalance via bladerf-cli thanks...

[bglod]

Please review the Wiki article on this topic. Please respond with any questions!

[ehhewng07]

I read it, but no information on how to calibrate IQ imbalance in auto way. Or can you provide how to use by bladerf-cli ?

[rtucker]

There's currently no built-in automatic method to perform I/Q calibration.

If you don't have a signal generator readily available, you can do an RX I/Q calibration using loopback: https://github.com/Nuand/bladeRF/wiki/D ... ctionality . You will want to do a TX I/Q calibration against a spectrum analyzer (or other known-good receiver) first, though.

[ehhewng07]

But, I have limeSDR, bladeRF,rtlsdr, it that possible to replace with spectrum analyzer?

[rtucker]

Your measurement results will only be as good as your reference equipment. Another SDR (or even the bladeRF's loopback) can be used as a spectrum analyzer for this purpose, but if its IQ/DC imbalances aren't known and compensated for, your results will not be ideal.

[ehhewng07]

Thanks, the calibration function works by bladeRF-cli, srslte have a bug, I know how to fixed, it start burst transmit twice without end transmission. Now, I can run srsLTE normally with bladeRF.

[Guillaume]

Hi,

I have exactly the same problem you had.
Trying to run srsue and srsenb on a bladeRF (x300 equiped).
I succeeded to run srsenb one month ago but no way to make it work again... I test it with a smartphone equipped with a green card accordingly programmed and with Nextepc as the core epc.
I never succeeded to run srsue... it detects mib, sib1, sib2 and sib3 (eventually the other sib), prepare the rach and then it seems that never it sends RA... and loops over sending rach with increasingly power...

I'm wondering if it could be a calibration problem...

Could you please send me the patch to fix srslte against the bug you talk about and your full procedure for calibrate the rx and tx bladeRF paths... (I do not have laboratory's tools)

Thank you for your help!
Regards,
Guillaume