OP25 audio still not processing

[MCWKen]

After setting up the desktop, and running OP25 in VirtualBox, the audio still is not totally being decoded. I have spent the better part of the week trying to tweak it referring to the many posts here.

When initiated, the TBSKS shows active until a voice channel is noted, the system tunes to that voice channel, and TBSKS stops.

Command Line:
./rx.py --args 'rtl' --gains 'lna:45' -T trunkmusc.tsv -S 960000 -X -x 1 -l http:127.0.0.1:8080 -2 -w -V -v 11 2> stderr.2

If I were to set -S to 250000, I will get a little bit more voice, but again it is not being processed thoroughly.

And advice to what needs to be changes would be appreciated.

 

[boatbod]

Issue #1 : you do not appear to have the audio player enabled. Add "-U" to the command line unless you are planning to stream to liquidsoap.

Issue #2 : that's a rather crappy low-quality signal you're pulling in. Notice the large "X" on the correlation plot rather than 4 nice tight corners of a square box, plus the smeared datascope plot. If I were you I'd look at improving your antenna, especially since you have the gain cranked up to max.

Issue #3 : there is no need to run a RTL wide open at 2.5Mhz sample rate. Try decreasing it to -S 1000000.

Note: the logs don't show any call traffic (complete absence of "voice update:" messages).

 

[MCWKen]

The antenna is a Larson NMO 800mhz with 55 foot of RG58a/u cable, 12 feet in the air. Best signal I can get from any of the five towers in the area (odd 774.33125 is the furthest from me; 30 miles - but the closest, 3 miles away is half the signal strength).

Set -S to 1000000 and this is the result.

I suppose if this is the most optimal settings, I will shelve this project til I can purchase a tower.

Thanks for your time.

 

[boatbod]

You might be encountering multipath - try a directional antenna. The signal should look more like this:

 

[KA1RBI]

I suppose if this is the most optimal settings, I will shelve this project til I can purchase a tower.

looks to me like there probably should be enough signal to work with - especially with the FFT showing more than 20 dB above the noise floor.

You should try setting the lna gain to 49, also try re-orienting the antenna until you get a small opening in the very center of the plot, at point (0,0). The fact that the constellation is sort of X shaped as opposed to four distinct points is no problem - as long as you can get the very center of it to be free of symbols. By moving the antenna you should have no trouble finding a spot where the center point becomes clear.

I'd also like to see what the symbol plot looks like.

Also, should be noted that it's normal for the tsbks counter to stop counting after switching to a voice channel...

Max

 

[pinkfish457]

Check out this website: https://www.rtl-sdr.com/ It has a video on setting up both op25 for phase 1 & 2.