P25 TDMA Control Channel decoding -- requesting help from experts

Status
Not open for further replies.

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
I apologize to developers who are expecting me to upload sensible raw audio. Much of the audio in the past two posts was taken with a 4.0k or 9.5k filter on it -- not good. Needs to be 12.5 or greater. I'll try to weed out the unsatisfactory in the future.

Mike
 

btt

Banned
Banned
Joined
Mar 11, 2020
Messages
2,585
Location
Wa State
The following is the valid data that I could get out of the 9.5kHz bw file. It did not have a TDMA_IDEN, so I assumed a base freq of 851000000. If that isn't correct, then the freq values are wrong.
WACN 0x91f14 SYS_ID 0x2d7 NAC 0x008
SITE_ID=8 RFSS=8

P25P2_CC----------------------------------------------------------------------
P25P2_CC NAC: 0x008, (dec) 8
P25P2_CC_FILENAME: Florida_Duke_Raw-Input_9.5khzBW_2021-07-24@155121.wav
P25P2_CC----------------------------------------------------------------------
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: RFSS_STS_BCST (0x3a): lra=8, sys_id=0x2d7, rfss=8, site_id=8, iden=8, ch1=1020 (freq=857.375000 MHz), srv_class=112
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: RFSS_STS_BCST (0x3a): lra=8, sys_id=0x2d7, rfss=8, site_id=8, iden=8, ch1=1020 (freq=857.375000 MHz), srv_class=112
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: SCCB (0x39): rfss=8, site_id=8, iden1=8, ch1=656 (855.100000 MHz), srv_class1=4, iden2=8, ch2=1492 (860.325000 MHz), srv_class2=4
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
$P25_PII_CC: NET_STS_BCST (0x3b): lra=8, wacn_id 0x91f14, sys_id=0x2d7, iden=8, ch=1020 (freq=857.375000 MHz), srv_class=112, nac=0x008
 

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
The following is the valid data that I could get out of the 9.5kHz bw file. It did not have a TDMA_IDEN, so I assumed a base freq of 851000000. If that isn't correct, then the freq values are wrong.
WACN 0x91f14 SYS_ID 0x2d7 NAC 0x008
SITE_ID=8 RFSS=8

Thank you. Base = 851.0125. That would turn 857.375 into 857.3875 (which was the freq monitored). Looks like this is going to be Duke Energy P25 out of Florida (nothing in DB yet, but @batdude figures it is Duke, and it certainly sounds logical to me).

m
 
  • Like
Reactions: btt

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
These should be the secondary CC frequencies then. 855.112500, 860.337500

Very nice -- Those two freqs and the original 857.3875 match the FCC license WRYH215 Ovieda location exactly. You nailed it. And it's definitely Duke Energy.

m
 

btt

Banned
Banned
Joined
Mar 11, 2020
Messages
2,585
Location
Wa State
Was able to get a better decode. Here are some adjacent sites.
$P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=12, freq=855.862500, iden=8, freq_off=4850000, rfss=12, sys_id=0x2d7
$P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=20, freq=860.387500, iden=8, freq_off=9375000, rfss=20, sys_id=0x2d7
$P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=31, freq=857.312500, iden=8, freq_off=6300000, rfss=31, sys_id=0x2d7
$P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=48, freq=856.387500, iden=8, freq_off=5375000, rfss=48, sys_id=0x2d7
 

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
Was able to get a better decode. Here are some adjacent sites.
$P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=12, freq=855.862500, iden=8, freq_off=4850000, rfss=12, sys_id=0x2d7
$P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=20, freq=860.387500, iden=8, freq_off=9375000, rfss=20, sys_id=0x2d7
$P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=31, freq=857.312500, iden=8, freq_off=6300000, rfss=31, sys_id=0x2d7
$P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=48, freq=856.387500, iden=8, freq_off=5375000, rfss=48, sys_id=0x2d7

Those peers would make sense in that I have a Google Map of all Duke site license locations with freqs. And on that map, there are locations very near site 8 Oveida that have those freqs on their license. So it's a good bet that the data is 100% correct above, at least as far as freqs go.
 

btt

Banned
Banned
Joined
Mar 11, 2020
Messages
2,585
Location
Wa State
With a change to the decode, the TDMA_IDEN and those adjacent sites all passed CRC, so it should all be correct.
 

btt

Banned
Banned
Joined
Mar 11, 2020
Messages
2,585
Location
Wa State
I don't know if this will add anything to what is already known about the information in the wav files, but I've optimized the demod for 4-fsk and re-run the decode. The summary includes the number of repeat messages for each opcode. It might be worth checking to see if any new adjacent or secondary sites were captured.
 

Attachments

  • complete_summary_optimized_demod.txt
    23.4 KB · Views: 12

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
I don't know if this will add anything to what is already known about the information in the wav files, but I've optimized the demod for 4-fsk and re-run the decode. The summary includes the number of repeat messages for each opcode. It might be worth checking to see if any new adjacent or secondary sites were captured.

I'll check it out. Thanks!
 

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
I don't know if this will add anything to what is already known about the information in the wav files, but I've optimized the demod for 4-fsk and re-run the decode. The summary includes the number of repeat messages for each opcode. It might be worth checking to see if any new adjacent or secondary sites were captured.

Everything is exactly the same CC/ACs/Peers on all of them, except that on Corpus (which is actually the Kenedy site) is listing _less_ peers than originally.

I believe originally the Corpus output listed:
8.8, 14.14, 41.41, 42.42, 43.43, 44.44, 114.114 peers

And your capture today from the Corpus audio lists:
8.8, 13.13, 42.42, 43.43, 114 peers

That is, unless I got confused on my end and documented something wrong. Actually for the "Corpus" recording I see descrepencies amongst many posts / multiple developers. I think we'll just have to wait until we have good raw audio to use from the get-go, and wait until ya'll have your ready-for-prime-time updates deployed to know for sure what the peers are on Corpus (Kenedy)

Mike
 
  • Like
Reactions: btt

btt

Banned
Banned
Joined
Mar 11, 2020
Messages
2,585
Location
Wa State
site_id = 14 is the primary Corpus (and is listed).

site_id 44 is an adjacent and has 3 records if I disable crc checks. One record for that looks like it could be correct and probably is since that is the GWest site. That explains that one.
1 $P25_PII_CC: ADJ_STS_BCST (0x3c): site_id=44, freq=854.762500, iden=8, freq_off=3762500, rfss=44, sys_id=0x1f8

I'm not able to find site_id 41 in the Corpus file, but it would make sense. This has been a lot of fun. I'll move on to getting support into the hardware. Thank you!
 

KA1RBI

Member
Joined
Aug 15, 2008
Messages
799
Location
Portage Escarpment
Just updating to note that rc2-final has been pushed to op25 osmocom master
Code:
git clone https://git.osmocom.org/op25

This includes an initial experimental release of the TDMA-CC code. See the README-July-2021 file for details. At the present time it should handle acquisition of basic control channel data (various IDs, channel map and neighbor list, etc).
 

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
New version of DSDPlus released. It copies the raw audio files and produces results.

 

AuggieActually

Member
Premium Subscriber
Joined
Oct 19, 2019
Messages
52
Location
Southwest BFE, OH
I have a few phase 2 CC samples of Duke Energy sites in southwest Ohio recorded with the new version of DSDPlus. Some are a bit noisy because my local MARCS site craps all over this portion of the spectrum.

I can try to get cleaner recordings from different locations if it's needed or desired.

1GB of IQ samples
 

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
10,368
Location
Carroll Co OH / EN90LN
I have a few phase 2 CC samples of Duke Energy sites in southwest Ohio recorded with the new version of DSDPlus. Some are a bit noisy because my local MARCS site craps all over this portion of the spectrum.

I can try to get cleaner recordings from different locations if it's needed or desired.

1GB of IQ samples

would you obtain a raw audio capture straight from DST plus by pressing R To record for 30 to 60 seconds and then pressing are to stop the recording? If you could, would you get one capture from a site that broadcasts neighbors but doesn’t broadcast alternate control channels, and then another from a site that does not broadcast neighbors but does broadcast alternate control channels.

thanks
 

AuggieActually

Member
Premium Subscriber
Joined
Oct 19, 2019
Messages
52
Location
Southwest BFE, OH
would you obtain a raw audio capture straight from DST plus by pressing R To record for 30 to 60 seconds and then pressing are to stop the recording? If you could, would you get one capture from a site that broadcasts neighbors but doesn’t broadcast alternate control channels, and then another from a site that does not broadcast neighbors but does broadcast alternate control channels.

thanks

All of that is included in the file. I grabbed a sample from each site I could pick up and decode a bit of.
 

rattlerbb01

TX/LA Database Admin
Database Admin
Joined
Dec 22, 2004
Messages
2,224
Location
Boerne, Texas
I'm new to this thread, but wow. As of two weeks ago the AEP sites in the western half of South Texas (Dilley, Bigfoot, Charlotte, Encinal, Uvalde) were all still EDACS analog. I'll be in the area for the rest of 2021 down there so if any of them convert (some used to list Freer et al as neighbors) I'll try and pass along whatever I can capture with SDR. It looks like DSD Plus will be a big help with the latest release.
 
Status
Not open for further replies.
Top