[USRP-users] decimation rate for gsm-receiver-usrp.py

Perper perper at o2.pl
Sat Sep 20 14:30:56 EDT 2014


W dniu 19.09.2014 o 22:37, Red Gator via USRP-users pisze:
> There is an interesting application called airprobe that allows an SDR
> to decode GSM signals.
>
> I have it mostly working with a B210, but it's acting like the
> decimation rate is wrong, and I am frankly clueless with regard to how
> to determine the appropriate deci rate.
>
> The RTL-SDR uses a decimation of 64, and the USRP1 uses 112.
>
> Can somebody suggest a decimation rate for the B210? 
>
>
>
>
> _______________________________________________
> USRP-users mailing list
> USRP-users at lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
Hi,

The gsm-receiver was written for USRP1 which had sampling frequency
64MHz. The sampling frequency is hardcoded somwhere in the constructor
of the flowgraph. The decim parameter was used to compute actual
sampling frequency coming out off USRP after decimation. As far as I
know B210 has arbitrarily variable sampling frequency (correct me if I'm
wrong), so instead of using decimation as a parameter you can set
directly sampling frequency.

The old gsm-receiver has problems with keeping synchronization and
proper frequency correction.

I'm working on new version outside of airprobe project, more modular,
with cmake buildsystem, and in which I've corrected these problems.
There are also examples for usrp and rtl-sdr that are working in
realtime. It doesn't have wireshark output as I didn't have time to make
it yet (maybe someone more experienced with wireshark want to help?),
but it can write to the standard output contents of the decoded messages
with use of separate block. If you want to give it a try I can give you
few hints how to start. The github page of the project:

https://github.com/ptrkrysik/gr-gsm

Best Regards,
Piotr Krysik




More information about the USRP-users mailing list