24 Feb
2013
24 Feb
'13
10:08 p.m.
Thinking about this some more, why couldn't you
just split the audio from the RX and TX into two
Yeah, asterisk app_rpt transmitting analog FM voice over standard FM radio hardware is
already a fully solved problem having no need for a USRP.
Our software stack can receive analog NBFM in addition to P25, for one thing because the
software support to do so was not difficult to add.
I thought the USRP_TX app was only for that hardware.
I want to use a conventional radio, just as you described, for both P25 and NBFM. So sign
me up as as experimenter wanting to try that!
OK -excellent. For starters could we
try first to get python/op25_audio_tx.py working - that app does not support asterisk /
app_rpt but will establish a baseline. The results of that test would direct the next
step of the experiment.
op25_audio_tx takes input analog voice either from a sound card or from a disk file (8,000
samples/sec, short signed int16 format). It generates audio at the soundcard audio
line-out port (which should be connected directly to the modulator stage of the FM TX).
You might be able to use the same sound card for both lines, if the card supports full
duplex and if it allows different rates for record and playback. If not the next best
thing might be to use canned test audio files [use the -i option and place the audio test
file in baseband-0.dat].
The biggest problem so far has been group delay distortion. It's also critical to get
the proper audio signal levels going into the TX such that the FM deviation is correct.
That should be it - let me know of next step(s)
73
Max