On Thu, 2 Apr 2015, Espectro Digital wrote: seems rtl_sdr is not generating the proper format. this was one of the items on my list, give me a few days - i will try to find a solution during the easter holiday, and send it to you. the software was meant to be used with multiple receivers, so this should not be a problem. [...] i used the python stuff because it was already there, but prefer c too. the qpsk demodulator is in python (i might try a c version later), the rest down the chain is proper c. jacek

On Thu, 2 Apr 2015, Espectro Digital wrote: sorry for my previous post, got the lists mixed up :) but since i've already answered, here goes: don't use rtl_sdr with low sample rates (<1Msps), either it doesn't work, or works bad. also tetra-demod.py will segfault in gri_mmse_fir_interpolator_cc() - and this is the probably the main reason for the error that you're seing. if you want to use rtl_sdr, then the easiest will be to write something that decimates the data to 36ksps and feed it to the qpsk demodulator (use simdemod2.py if you don't want to write your own). this way most of the processing chain will be C (the qpsk demodulator can be exchanged later for a c version too, but i doubt this will significantly lower the system load). this is on my (long) todo list for telive, because this would allow using cheap hardware for listening to a lot of channels at once. but it depends what you want to use rtl_sdr for. do you want it for less system load, or do you want it to have remote control over the receiver. in the latter case you can control gnuradio via http. i can post a gnuradio flowgraph that i use for scanning tetra networks. the frequency is controlled via http from a modified version of telive (scans the frequencies and logs the network parameters it finds) jacek