PSERN + Whistler TRX-1 issue

[devicelab]

Lately, my PSERN decodes on the Whistler TRX-1 were very poor at best. I would get a 30s recording and only about 3-4s would be actual voice. This happened over and over. Some TGIDs seem to fair better than others. At first, I thought it was a specific user but then I noticed it happening system wide. In terms of a percentage, I'd say that it was down to about 33% receive quality.

I think this may be a symptom of the LSM Distortion which is inherent in the TRX scanners -- but this is the first time I've experienced it. My SDS-100 and Airspy R2/DSD+ can decode PSERN pretty much flawlessly. (I reported major issues with an older X2-TDMA system many years ago but that issue went away when they upgraded to Phase 2.)

I found one setting that seems to have helped the quality from 33% to about 85% -- I doubled the "Dwell Time" under the PSERN system settings.

I'm curious if anyone else has had similar issues lately.

 

[Ubbe]

I found one setting that seems to have helped the quality from 33% to about 85% -- I doubled the "Dwell Time" under the PSERN system settings.

Dwell are the time the scanner stays at the site after it has lost its decoding of a voice call. So for sure it is a simulcast issue that makes the scanner loose the signal intermittently, if you still have a good signal strength. You do the same thing with a Uniden scanner to set a longer hold time, that are usually called dwell in other scanners and radios, to let the scanner try for a longer time to decode the signal before it gives up and leaves the system.

A radiowave are about one foot long and half that distance will go from max to min so mowing the antenna just 1/2 foot can make a difference. The frequency drift of the scanner can also make a difference when receiving simulcast systems. And then also any changes they can have made on the systems antennas and power outputs can make a change how well you can receive the system with a non simulcast receiver.

What you can try are the DSP level adapt and ADC gain settings. They will have an impact on how a P25 digital signal are handled. The DSP you can try and change in 5 steps increments and see if it improves anything and the DAC in 2 dB increments.

/Ubbe

 

[devicelab]

What you can try are the DSP level adapt and ADC gain settings. They will have an impact on how a P25 digital signal are handled. The DSP you can try and change in 5 steps increments and see if it improves anything and the DAC in 2 dB increments.

I ran DSP @ 48 and -2/-6 because my TRX is always at home on an external antenna. (The P25 audio quality sounded better with a lower DSP.) When I started noticing issues, I reverted back to the defaults. It didn't change anything good or bad, so I didn't bother to continue mucking with it. I'm OK with this for now.

 

[Ubbe]

I ran DSP @ 48 and -2/-6 because my TRX is always at home on an external antenna. (The P25 audio quality sounded better with a lower DSP.) When I started noticing issues, I reverted back to the defaults. It didn't change anything good or bad, so I didn't bother to continue mucking with it. I'm OK with this for now.

-6dB are the audio level to the speaker for all digital systems, to match the audio level to systems with analog signals and are not related to the decode process. The default -2dB are the one that matters for P25 decoding quality, but does nothing for DMR/NXDN.

If you have a good signal strength from the system, 3 bars or more, you can try attenuating the signal to try and receive just one tower and make the others disappear below the receive sensitivity of the scanner. You can get a kit with inline attenuators of 3-6-10dB to try in different combinations and some adapters to fit your coax.

/Ubbe

 

[devicelab]

If you have a good signal strength from the system, 3 bars or more, you can try attenuating the signal to try and receive just one tower and make the others disappear below the receive sensitivity of the scanner.

Just to be clear, I'm only monitoring the primary simulcast site -- which is about 10 cells -- and yes, the signal is very strong.

 

[devicelab]

Quick update: I increased the Dwell Time to 2.0 secs and also adjusted my DSP back down to 48. The quality seems slightly improved.