Filter Decoding on System ID?

[Ben Saladino]

Is there a way to limit control channel decoding to a system ID?  I have an alternate control channel on a P25P1 system that picks up a P25P2 system that is close in frequency. I'd like to be able to filter that out.

[wemana...@gmail.com]

On Friday, December 24, 2021 at 8:51:36 AM UTC-5 ben.sa...@gmail.com wrote:

Is there a way to limit control channel decoding to a system ID?  I have an alternate control channel on a P25P1 system that picks up a P25P2 system that is close in frequency. I'd like to be able to filter that out.

I am going to take a stab at this, and this may possibly be related to an issue I've had for a while in regards to something else...

What you are describing in the real physical radio world would be adjacent channel interference aka selectivity. Which causes "bleed over." I don't want to go down a long boring post on stuff of RF, but simply lets just say you have 853.850 as a CC and 853.875 as a CC as well. Two different systems, the selectivity of the radio/SDR is what determines how well these work. In practice it should filter these out when tuned to either channel. 

So when you do this on a real physical radio scanner or actual APX etc. then the filters actual little filters in the radio filter out things. This is normally done at the IF section of the radio. Hence why scanners, early ones, had issues with these using cheap filters... SDR would have something similar but its done via software, and actually SDR programs should be able to do this quite a bit better v. a physical filter. As some of the issues in a physical filter v. a "virtual" SDR filter won't exist. Now where this come into an issue in SDR is COMPUTING POWER. Lots of fancy math involved in this.. stuff... that just barely existed when I started all this... and not my thing.

I will give an example: To do the NOAA NWR feed, I use my strong "local" signal. Now because I have a good setup, and NWR's sites are putting out 1KW on each of these sites, and nice mostly flat terrain I can pick up numerous other sites. Some 50+ miles away. Some even have multiple sites squashing each other. Especially in a tropo event. Using just the standard rtl_fm and the cheapo asian RTL and even the "V3" trying to monitor the CLOSE ADJACENT channels on NWR on the SDR is terrible v. a scanner or LMR which nicely discerns the channel. I found through some experimenting that using the -F fir_size and setting it to like 4-5 (this test was done awhile ago, going on memory, I didn't/don't use in production) allowed the close channels to be received with out the bleedover of the adjacent channels.  Ie: my local 162.550 would bleed through on just about everything +- 50khz or so. The next closest 162.500 actually would be garbled on the SDRs but on  scanner fine to get the distant station. Same for some on other more distant that might have an adjacent next to them. The FIR option cleaned it up to perform "correctly."  :) ;)

So what I am getting to is that I think there might be some chances to improve the FFT operations to filter out close and other interference in the sampling to fix this problem. I have a transient RFI situation wherein some stuff interferes with one system. Physical radios don't have this issue, BUT SDRT and OP25 do have a problem with this. 90% of the time its no issue. I could list all the things I've done on this to chase this, but I don't want to bore you or others, and its kinda infuriating this issue...

So my question to you would be how close are these channels, because technically, if you are in say county a, county b should NOT have an adjacent channel to county a and b! If they do then you have poor regional plan directors ( :) ;)  ) How close are these channels? How much have you optimized RX? Ie: LNA's? Yagi antennas pointed into the same direction?? etc.. Additionally tropo aka skip can occur for even 700 and 850Mhz systems. In the right conditions I've actually gotten sites half way across the state! even on a portable.

The point to this is that the way the FFT's is done to slice up this and then decode it, might need some "fine tuning" to be more selective versus STRONG adjacent or near adjacent signals. Denny would have to comment on that... but I just taking a guess on translating from real radio to SDR on how this would work, as this FFT stuff is not my thing.
 

[Ben Saladino]

I realize the inexpensive and wide ranging SDRs aren't going to be able to filter out everything. I'm hoping for a way using the NAC or system ID to make sure that only the match system traffic and control channels get through.