P25RX-II P25RX-II Firmware Updates / Testing

[MStep]

I certainly don't speak for Todd, but before lots of folks (including myself) start heaping on a bunch of suggestions, especially for an update which hasn't been officially released yet, I think we should all be patient and see with the new release is going to look like. Once that is stabilized would probably be a better time. I imagine there will be some "bug fixes" that with have to be done first, as it should be. From my own point of view, it's probably the most logical way to proceed. One man's opinion. With the usual FWIW and YMMV thrown in for good measure.

 

[btt]

Todd,
Can the P25RXII have the "allow unknown TG" turned on by system or is it all on or all off? Maybe a nice feature to have if its all on all off. Just a thought (like you need more to do, Sorry)

I hadn't thought of that, but I think that would be an easy thing to do and shouldn't cause any issues that I can think of.

 

[kb8q]

I hadn't thought of that, but I think that would be an easy thing to do and shouldn't cause any issues that I can think of.

I guess I don't understand the suggestion since I'm doing that with the current software. 3,000 TGID's in the editor spread across 9 zones. The ones I don't want to hear normally are in zones 1 & 2 which I keep disabled. The config allows unknown TG's which I want to receive because they're unknown. If I don't want to discover anything I disable the ability to receive TG's not in flash. I periodically add the unknowns to flash into appropriate zones based on whether I want to hear them again. Once they're in flash, they won't show up as unknown again even if they're disabled in the editor.

 

[Mike_G_D]

I guess I don't understand the suggestion since I'm doing that with the current software. 3,000 TGID's in the editor spread across 9 zones. The ones I don't want to hear normally are in zones 1 & 2 which I keep disabled. The config allows unknown TG's which I want to receive because they're unknown. If I don't want to discover anything I disable the ability to receive TG's not in flash. I periodically add the unknowns to flash into appropriate zones based on whether I want to hear them again. Once they're in flash, they won't show up as unknown again even if they're disabled in the editor.

I believe that what was suggested was to have the "Allow Unknown TG's" option to be tied to each system separately so that one could have unknowns allowed for one or more systems that are being scanned but not for others. For example, I have a major local county system and a near by adjacent (different) county system. So, I might want to allow unknown's on the local county system but not on the adjacent county system just to help me keep things "clean" during my human-limited sorting ability (since it is possible that two completely un-related systems can use the same TG's to mean different things you really want to keep them separate when looking at multiple systems and scaning for unknown TG's).

 

[EMwaves54]

Anyone else stalking the forum waiting for the big reveal?

 

[btt]

Anyone else stalking the forum waiting for the big reveal?

Been working on it since this AM. I'm currently stuck on an issue with writing the channel configs to flash. Once I get that figured out, it shouldn't take long to get something ready for a testing release.

 

[btt]

Finally got past the writing channel configs structures to flash. It was a compiler thing adding extra bytes for padding. I forgot __attribute__((packed))

 

[dcr_inc]

Mike_G_D... Exactly what I would like to do..
Todd,
You are the man !!.. Impatiently awaiting the latest and greatest update !!!

 

[btt]

This is a link to the new experimental release page. P25RX-II experimental firmware
I have tested this version for about 10 minutes. It is NOT complete. I will add notes to the above link a bit later today. Please don't download this if you are not in the mood for incomplete software. This release is completely different than previous releases. One large change some of you will notice right away is that there is no longer a "SearchDB tab". The integrated P25 zip-search DB has been removed. This will become a separate application with optional, downloadable database files. This will allow supporting systems other than P25. There is a new fast, signal-search feature. Once you figure out how to use it, you may find that this works well enough that you don't need the zip-search!

A few random notes:
Scanning is disabled on power-up. You have to manually enable it with the "SCAN" button for now. (this will change)
When not scanning, you can double-click on a channel to change to that frequency.
For optimal scanning, you may want to select fixed-gain settings for each scanned channel. You can do this per channel.
The mouse scroll-wheel works on "CH STEP", "FREQUENCY DISPLAY", "SQUELCH".
After you make a change to a channel setting, press the "APPLY CHANGES" button to keep your changes. This will save ALL configured channels to a file called "channel_config". For scanning, you will need to transfer the channel config information to the P25RX-II with the "SYNC Channel Config To Receiver" button.
DMR Trunking is not implemented in this release. You should be able to scan / monitor DMR channels in "conventional" mode.
Signal Insights, Console, and Display view tabs all support popping-out to a separate window now.

Signal Search:
1) Turn off scanning
2) Set the squelch level with the slider
3) Select band
3) Choose either P25-CQPSK with 100ms dwell for digital detect or NBFM with 20ms dwell (prefer this for speed!). If you use the NBFM, you can still configure detected channels for digital after the search is complete/aborted.

The P25 Config and DMR Config tabs under the Channel Config tab are NOT implemented yet.

Now that the hard stuff is done, there should be a lot of progress for features over the next week or two. I need to make more videos and write a lot more documentation. Hopefully, it is in good enough shape for those of you who are feeling adventurous and want to play with the new features. It has been working excellent as I type this. I will work on this some more this afternoon, but will be busy for a good part of tomorrow working on hardware. Have fun!

 

[btt]

Already found 2 bugs. 1) You have to definite both trunked and non-trunked channels for scanning to work. 2) When scanning, it misses the last channel in the list for both trunked and analog. To get around this for now, define an extra channel enabled, in the scan, and set the squelch very high, e.g. 0. This last channel in the list will be skipped. This needs to be done for both a trunked and non-trunked channel. I will get these bugs fixed later this evening.

 

[freqseeker]

Looking forward to trying it out. Should be able to devote some time to it late tomorrow. I'll post my results.
As usual thank you for all your hard work.
Vince.

 

[btt]

The bugs mentioned in the previous post have been fixed in version 2022-04-23_1443.

 

[btt]

For those of you checking out the new software / firmware, I thought I would point something out regarding LSM (linear simulcast modulation). I've posted some things in the past that implied the main issue with simulcast for traditional receivers is the delay spread. The more I've learned about LSM, I've come to the realization that delay spread is most likely not the main issue for older receivers. As you may know, a traditional FM (frequency modulation) signal has a constant amplitude. Watching the IQ plot, with the NBFM demod, you will notice that FM signals mostly look like a circular plot. For a constant carrier, you may see a dot somewhere on the circle, or CW / CCW rotation for +/- phase. The constant amplitude means it will basically stay on the circle and only the phase will change. Now, monitor some LSM vs non-LSM P25 control channels with the NBFM demod. You will notice that the amplitude is NOT constant for the LSM sites. This is the main reason that receivers with traditional FM detectors have trouble receiving LSM. The delay spread from multiple towers can be an issue, but only if you are receiving strong signals from towers that are MANY miles apart. It isn't very likely for most circumstances. You can try it for yourself with the latest release. The other thing to notice is that while LSM and non-LSM are both deferentially encoded, the differential symbol phase remains much tighter for the LSM sites. With quadrature signals, DSP, and the right demod, the LSM signals are actually preferable!

 

[maus92]

Thank you for your continued development on this radio project. TBH, this radio is not the best performer on the systems I monitor using the current f/w release, but I decided to keep the radio based on the promise of future capabilities. Trying to find the time today to reflash the radio.

 

[btt]

Thank you for your continued development on this radio project. TBH, this radio is not the best performer on the systems I monitor using the current f/w release, but I decided to keep the radio based on the promise of future capabilities. Trying to find the time today to reflash the radio.

I thought it was working well for you in the past. Do you know what version quit working well? I've been shipping with 02-10_1045 stable. I'm not aware of any performance issues with it. email me if you think it could be a hardware issue. I will take care of it.

 

[maus92]

TBH, I don't remember which version was working the best for me (maybe 2-25 or 2-10,) but I do notice that clear, calm days - like today - show the best performance. On cloudy, rainy, windy days, I get widely variable RSSI with attendant voice artifacts that you would expect with fluctuation signal power. My other receivers don't seem to be as affected, if at all, by wx. Here are some screenshots of RSSI expressed in terms of dBm - although who knows how accurate they actually are... The other programs are UT2 and SDRTrunk. The RSSI on a SDS200 is -60dBm. Note that the BT radio is decoding a voice channel which is usually lower in power; today the RSSI on the cch is ~-85dBm. All are connected to the same antenna system, and I've ruled out bad cables.

 

[btt]

@maus92,
Try setting the AGC step time 350ms and the AGC target to -20. This works well for another person that was having some similar issues. With the new firmware/software, the gain settings can be configured per frequency. You may find this useful for fixed location monitoring on your shared antenna setup and the variance in power for different channels.

 

[btt]

There was still an issue with P25 P2 when scanning is enabled. I think I found it this time, but have not verified the fix yet. Version 2022-04-24_1213 is available.

 

[FreqNout]

For those of you checking out the new software / firmware, I thought I would point something out regarding LSM (linear simulcast modulation). I've posted some things in the past that implied the main issue with simulcast for traditional receivers is the delay spread. The more I've learned about LSM, I've come to the realization that delay spread is most likely not the main issue for older receivers. As you may know, a traditional FM (frequency modulation) signal has a constant amplitude. Watching the IQ plot, with the NBFM demod (raw I/Q plot), you will notice that FM signals mostly look like a circular plot. For a constant carrier, you may see a dot somewhere on the circle, or CW / CCW rotation for +/- phase. The constant amplitude means it will basically stay on the circle and only the phase will change. Now, monitor some LSM vs non-LSM P25 control channels with the NBFM demod (this will plot the I/Q without a demod). You will notice that the amplitude is NOT constant for the LSM sites. I believe this is the main reason that receivers with traditional FM detectors have trouble receiving LSM. The delay spread from multiple towers can be an issue, but only if you are receiving strong signals from towers that are MANY miles apart. It isn't very likely for most circumstances. You can try it for yourself with the latest release. The other thing to notice is that while LSM and non-LSM are both deferentially encoded, the differential symbol phase remains much tighter for the LSM sites (like coherent phase QPSK). With quadrature signals, DSP, and the right demod, the LSM signals are actually preferable!

Great info on the IQ plot. It has been a long time since I studied complex numbers and polar plots. It took a few reads to follow it all. In some of your future p25rx documentation,perhaps a few pictures explaining what the IQ plots are showing to help the non engineers and us old engineers that need a refresh.

 

[btt]

Great info on the IQ plot. It has been a long time since I studied complex numbers and polar plots. It took a few reads to follow it all. In some of your future p25rx documentation,perhaps a few pictures explaining what the IQ plots are showing to help the non engineers and us old engineers that need a refresh.

I've been thinking about doing something like this. It is actually much easier to understand with 3D perspective plots for the quadrature signal over time. Really cool for an OFDM signal! I'll work on it.