bogdandia / airprobe-hopping Goto Github PK

Hello,

I have captured a cfile using grgsm_capture: grgsm_capture.py -a 79 -g 98 -s 2e6 -c out/out.cf
I used channelize script to channelise the wideband file. From System Information 1, the list of arfcns are 59,77,79. My main arfcn is 79. MAIO:0 HSN:0. Obtained MA from Immediate assignment.
I edited the script with following configuration:

CONFIGURATION="1S"
CA="59 77 79"
C0="79"
MA="07"
MAIO=0
HSN=0
KEY="00 00 00 00 00 00 00 00"

But I do not see any output in wireshark.
Can you please give me some suggestions or inputs to get this to work?

Thank You

You can adapt your package of files to the version gnuradio_3.8.0.0 ?

Hi,
I have been trying to get this code working for me for more than a month now, so I'm pretty confident I know my way around it, so I would like to give some input and get some help:

1. I captured many files using a USRP B200, and I was able to figure out what parameters to give to the channelize2.py script, so now I have all the channels split up for me in files. I checked how the cell works, and I distinguished between single hop (change to a different frequency for voice, but only once) and multiple hop (keeps changing frequency for each frame) cells. I know the code is for multiple hop, so I tried to decode such traffic, but it fails to give me voice data. The cell uses EFR, so the configuration of the receiver is TE, not T, but other than that it is just a hopping cell, so I really don't know why it doesn't work. I can maybe give you my capture files so you can give it a try maybe.
   I also tried to patch the code for single hop cells, (by specifying the cai value by hand instead of calculating it using the calculate_ca_sfh function), but also that fails. I'm always getting parity errors, so I tried disabling parity checks, but that resulted simply in a garbage speech file of course.
2. For GnuRadio 3.7 you need to change the center frequency in gsm_receive100_channelize.py to look like this:
   self.top_block.set_center_frequency(-freq_offset)
   (notice the negative sign in front of freq_offset)

If you don't do that the code will simply not work at all (not even with the broadcast channel) according to my experiments.

I would really like to get your ideas and opinion on these issues if possible. Thanks.

Hi Bogdan,

Excellent tool and guide you've written. I have recorded a "wideband capture" with a BladeRF using a slightly modified version of grgsm_capture, and I am able to "channelize" each ARFCN using grgsm_channelize.py.

I have tested that I am able to decode the individual files with grgsm_decode as well which works.

However, when I try to use your script (after modifying the parameters), with 0 decimation or 64 decimation, it doesn't work. I know that Airprobe for some reason, does not decode GSM frames/packets as well as gr-gsm for unknown reasons. (I only get some of the packets in wireshark with Airprobe.)

The combined capture/complex file created by airprobe-hopping is empty in my case, meaning the program that decodes the individual out_XX.cfile files, does not seem to understand the format I'm using. I tried modifying the sample rate, so that in my case I had nine ARFCNs at 1MSPS each, meaning I made the total MSPS to 9MSPS instead of the default which is 10MSPS.

I was wondering if you have any documentation about how you've implemented the frequency hopping into Airprobe? I'm not a programmer per se, but I am willing to try and implement this into grgsm for example if the official gr-gsm does not want to.

Let me know if you can help me out, I have figured out everything else so far, but I'm stuck on the frequency hopping part.

In my case, I have an additional obstacle at the "SDCCH8" part where frequency hopping is enabled, as my phone hops to another ARFCN, and then does the SDCCH8 part, and then hops back, but that's not really an issue for my testing at the moment, as I already have the Kc extracted.

g++ error ./gsm.cc no input
How can I solve it make file gsm-receiver