New TETRA decoding plugin for sdr# problem

Joined
Oct 6, 2017
Messages
21
#1
Hi,

Saw this new TETRA decoding plugin for sdr# "https://www.rtl-sdr.com/tetra-decoder-plugin-for-sdr-now-available/".
The plugin seems to work just fine and bursts are received in the MSYS2 window.
However,for some reason the newtork is not recognized properly and I can't hear anything, although I see broadcasts and traffic on the channels via the MSYS2 window.
I know that encryption is not the problem because the "Network Info" window is completely empty (parameters "Air encryption 1" or "Air encryption 2" should appear in the Network Info window, among other network data).
I see that the first CRC COMP says "ok" but the second says "wrong" in the MSYS2 window so that might be the reason, not sure.
Tried set "minOutputSampleRate" value in "SDRSharp.exe.Config" file to 32000 and use WFM / set to 48000 with NFM. In both cases windows playback set to 48000.
Still the same.
Once in a while I hear a very short sound (last less than a second), might indicating tunning error? not sure.
Posted pictures below.
Any ideas?
 

Attachments

Last edited:

mtindor

OH/WV DB Admin
Database Admin
Joined
Dec 5, 2006
Messages
7,210
Location
Carroll Co OH / EN90LN
#2
Hi,

Saw this new TETRA decoding plugin for sdr# "https://www.rtl-sdr.com/tetra-decoder-plugin-for-sdr-now-available/".
The plugin seems to work just fine and bursts are received in the MSYS2 window.
However,for some reason the newtork is not recognized properly and I can't hear anything, although I see broadcasts and traffic on the channels via the MSYS2 window.
I know that encryption is not the problem because the "Network Info" window is completely empty (parameters "Air encryption 1" or "Air encryption 2" should appear in the Network Info window, among other network data).
I see that the first CRC COMP says "ok" but the second says "wrong" in the MSYS2 window so that might be the reason, not sure.
Tried set "minOutputSampleRate" value in "SDRSharp.exe.Config" file to 32000 and use WFM / set to 48000 with NFM. In both cases windows playback set to 48000.
Still the same.
Once in a while I hear a very short sound (last less than a second), might indicating tunning error? not sure.
Posted pictures below.
Any ideas?
One idea would be to contact the author of the plugin. Just a suggestion :)

mike
 
Joined
Jul 7, 2011
Messages
120
#3
I also tried it and it worked fine. One thing that comes to my mind is that this channel is broadcasting data only. Other than that, I saw on your screenshot that you are getting only bursts in MSYS. On the network I tried I'm seeing network info also. Don't know why is this but that's the first thing I saw your post.
 
Joined
Oct 6, 2017
Messages
21
#4
One idea would be to contact the author of the plugin. Just a suggestion :)

mike
Wish I could, but the author of the plugin is unknown.
From rtl-sdr website:
"The plugin doesn't seem to be officially released anywhere, but we did find it thanks to @aborgnino's tweets on Twitter, and he found it on a Russian language radio scanner forum".
There is no github repository either.
 
Joined
Oct 6, 2017
Messages
21
#7
I also tried it and it worked fine. One thing that comes to my mind is that this channel is broadcasting data only. Other than that, I saw on your screenshot that you are getting only bursts in MSYS. On the network I tried I'm seeing network info also. Don't know why is this but that's the first thing I saw your post.
Yes I know the Network Info window should be filled with different network parameters, even if the network is encrypted.
That's why I know encryption is not an issue here, but something else.
Regarding data channels, I have lots of TETRA in my area and for all of them I have this problem.
It highly unlikely that all of them are data channels and some must be voice channels.
What version of SDR# and vc++ you use?

I wonder if Filter Audio in SDRSharp should be un-checked?
Tried both checked and unchecked.
In checked I hear short sounds often, still there is no such a big difference.
 

Ubbe

Member
Joined
Sep 8, 2006
Messages
2,556
Location
Stockholm, Sweden
#9
It looks as if you decode something as the Netinfo text is red that indicate it decodes info and also channels seem to populate with call numbers.
The only thing is the error value 0.21 at the polarity invert that seems a bit high, and that you monitor 900MHz but upload/download frequency are decoded as being in the 400MHz range?
Are you spot on the frequency, you have calibrated the SDR sticks frequency?
I only decode properly if the SDR stick sample rate are set at 2.4MSPS in the cog wheel config.

/Ubbe
 
Joined
Oct 6, 2017
Messages
21
#10
It looks as if you decode something as the Netinfo text is red that indicate it decodes info and also channels seem to populate with call numbers.
The only thing is the error value 0.21 at the polarity invert that seems a bit high, and that you monitor 900MHz but upload/download frequency are decoded as being in the 400MHz range?
Are you spot on the frequency, you have calibrated the SDR sticks frequency?
I only decode properly if the SDR stick sample rate are set at 2.4MSPS in the cog wheel config.

/Ubbe
Just updated to latest version 1637 of sdr# to see if that chaneged something. No luck so far :(
The error now is lower, between (+-)0.05 to (+-)0.11 values.
The "NETINFO" word (top left of the net info window) is red and indicates some unclear data.
Can you figure out if that a control / voice channel?
Not sure I understood your questions.
The TETRA channels freqs are at 933 - 938Mhz in my area.
Is there any difference regarding the decoding in these freqs comparing to the 400Mhz freqs?
I have calibrated my sdr sticks to that frequency manually, set the bandwidth to 25Khz and use WFM mode with 32000Hz sample rate in the "minOutputSampleRate" value at "SDRSharp.exe.Config" file.
Using 2.4 Msps in the cog wheel config.
Missed anything?
Posting below the updated pictures.
 

Attachments

Ubbe

Member
Joined
Sep 8, 2006
Messages
2,556
Location
Stockholm, Sweden
#11
It's coded to default to the 400MHz band, the only one used in EU, but that should only make the info about the control channel frequency being displayed incorrectly and probably not make any difference to the decoding.

But didn't the instruction point out that you need to use 48kbs audio stream?
I can see that #SDRSharp sends audio to msys in a rate varying between 32kbs and 44kbs so limiting it to 32kbs migh be an issue.

/Ubbe
 
Joined
Oct 6, 2017
Messages
21
#12
It's coded to default to the 400MHz band, the only one used in EU, but that should only make the info about the control channel frequency being displayed incorrectly and probably not make any difference to the decoding.

But didn't the instruction point out that you need to use 48kbs audio stream?
I can see that #SDRSharp sends audio to msys in a rate varying between 32kbs and 44kbs so limiting it to 32kbs migh be an issue.

/Ubbe
Well, the original instructions indeed pointed out that for WFM one need to use 32KHz output audio stream and for NFM at least 48Khz stream.
I have tried to use output stream values of up to 96Khz, because in my area the TETRA freqs are around 935Mhz.
Still no difference.
I am not sure what is the right output sampling stream for these high freqs or the sdr# upper limit.
In your case, what is the rate of output data sent to the Tetra-Rx?
Here it's between 32KB/s - 40 KB/s which is 256kb/s - 320 kb/s, so maybe the 96Khz output sampling sent to Tetra-Rx is not high enougth.
Unfotunately, trying the next step of 176.4Khz results in an error message by sdr#.
Once in a while I hear sounds when tunning to a TETRA channel, but it's too short to determine if it's a human or other metallic sound.
That's why I think the problem is has indeed something to to with the sample rates error (too low sample rate) or wrong demodulation type by the plugin, although I am not sure if there are different types of demodulation in TETRA based systems, like for example, in P25 based ones.
 

Ubbe

Member
Joined
Sep 8, 2006
Messages
2,556
Location
Stockholm, Sweden
#14
Tetra standard should be the same regardless of frequency band the only difference being 45MHz between tx/rx instead of 10MHz.

I also see the "CRC comp=wrong" message in the telgrams that say "ul_usage traffic" and must be the result of some coding error by the author.

The NFM WFM settings probably doesn't matter, you set the bandwidth yourself to whatever you want. It is only different memory settings for the parameters. I can use both without seeing any difference between them.

Have you tried all available tetra frequencies to see that this one isn't just some sort of special one for test during setup of the system or transfer of data only without voice?

All indicators seems to be ok in your picture but you never hear voice.

I use sample rate 2.4MSPS Quadrature sampling in SDR# for the SDR stick with audio sample 48000 sample/sec and for some reason I have 100mS latency, no filter.

Radio uses WFM with Blackman-Harris 4 filter.

/Ubbe
 
Joined
Oct 6, 2017
Messages
21
#15
Just in case, take a look around 420-428 MHz.
Tetra standard should be the same regardless of frequency band the only difference being 45MHz between tx/rx instead of 10MHz.

I also see the "CRC comp=wrong" message in the telgrams that say "ul_usage traffic" and must be the result of some coding error by the author.

The NFM WFM settings probably doesn't matter, you set the bandwidth yourself to whatever you want. It is only different memory settings for the parameters. I can use both without seeing any difference between them.

Have you tried all available tetra frequencies to see that this one isn't just some sort of special one for test during setup of the system or transfer of data only without voice?

All indicators seems to be ok in your picture but you never hear voice.

I use sample rate 2.4MSPS Quadrature sampling in SDR# for the SDR stick with audio sample 48000 sample/sec and for some reason I have 100mS latency, no filter.

Radio uses WFM with Blackman-Harris 4 filter.

/Ubbe
Well, thank you guys, that was finally it.
Searched carefully around 390-420 Mhz with better antenna and found 2 TETRA networks that did not notice before.
One of them (around 393Mhz) decoded properly by the plugin with minOutputSampleRate value of 32000, audio sample rate in sdrSharp of 48000, 44100 in windows output audio settings, WFM and 2.4 MSPS.
Heard voice properly on that one.
Trying to decode the second (around 420 Mhz) results in garbled metallic sound.
Thought of an encryption in the second network but then noticed both networks has the "Air encryption" field set to 1 from what I see in MSYS terminal, and the first newtwork is decoded properly, so not sure encryption is the issue with the second network.
Maybe you have a better idea what's wrong here.
Regarding the original TETRA network (around 933-938 Mhz) I tried to decode, searched what TETRA system might that be with no luck .
Online sources reports of the following bands for TETRA based systems:
410-430 MHz, 870-876 MHz, 915-921 MHz, 450-470 MHz, 385-390 MHz , 395-399,9 MHz of the ETSI standarts.
However, this network was definitely TETRA based one that the plugin recognize but can't decode for some reason.
Have and idea of what that reason might be?
Thought of a new TETRA network that supports/combine 4G/LTE features but not sure about that.
Would be glad to figure that one out.
If so, can telive (https://github.com/sq5bpf/telive) decode that?
Thanks again for the help.
 

Ubbe

Member
Joined
Sep 8, 2006
Messages
2,556
Location
Stockholm, Sweden
#16
I'm beginning to believe it is only some sort of image frequency you are trying to decode.

Your ETSI source says 915-921MHz and duplex -45MHz is 870-876MHz but you have a signal at 938MHz.
But you also have a signal at 393MHz and the nationwide system I have uses 390-394MHz so the source have omitted that frequency range, so maybe the source are not to be trusted.

For high UHF band I suggest you only search in the 915-921MHz range, there shouldn't be any non standard systems out there. For low UHF the mobiles transmit at -10MHz so look in the upper 10MHz part of each band, 420-430MHz 460-470MHz 390-399MHz.

/Ubbe
 
Joined
Jul 7, 2011
Messages
120
#17
Just what Ubbe said. This is a "ghost" signal if I can call it like that. That's about 915-921 MHz signal.

About encryption - Air encryption = 1 means that network is able (and most probably do) encryption but this does not mean that all subscribers use encryption during calls. That's why you are able to hear voice even if air encryption = 1.
 
Joined
Oct 6, 2017
Messages
21
#18
I'm beginning to believe it is only some sort of image frequency you are trying to decode.

Your ETSI source says 915-921MHz and duplex -45MHz is 870-876MHz but you have a signal at 938MHz.
But you also have a signal at 393MHz and the nationwide system I have uses 390-394MHz so the source have omitted that frequency range, so maybe the source are not to be trusted.

For high UHF band I suggest you only search in the 915-921MHz range, there shouldn't be any non standard systems out there. For low UHF the mobiles transmit at -10MHz so look in the upper 10MHz part of each band, 420-430MHz 460-470MHz 390-399MHz.

/Ubbe
Just what Ubbe said. This is a "ghost" signal if I can call it like that. That's about 915-921 MHz signal.

About encryption - Air encryption = 1 means that network is able (and most probably do) encryption but this does not mean that all subscribers use encryption during calls. That's why you are able to hear voice even if air encryption = 1.

A "ghost" signal is definitely an option here.
Made some further research and apparently there are countries with TETRA systems of 915 - 933 Mhz, which could explain these wierd freqs, still considering a "real" signal.
To take into consideration the SDR gaps and a chance that the one I used fails to point the right freq, which could lead to "ghost" signals, checked this again with 3 different SDR sticks (of different manufacturers).
All of them has these wierd TETRA freqs appear. Strange.

Anyway, the plugin should still be able to decode a "ghost" TETRA signal if it was "transferred" from 915 - 921 Mhz to 933 - 938 Mhz band, shouldn't it?
This in case of a stantard system of couse.

Regarding the encryption, thanks for info.

Btw, what is is the difference between this new plugin to telive regarding aspects of sound quality and decoding none standart or additional types of systems?

Thnaks.
 
Joined
Sep 19, 2002
Messages
4,941
Location
Toronto, Ontario
#19
Anyway, the plugin should still be able to decode a "ghost" TETRA signal if it was "transferred" from 915 - 921 Mhz to 933 - 938 Mhz band, shouldn't it?
Not necessarily. The polarity of the extracted signal could be inverted and if the decoder does not expect that, it won't handle it. Also, the imaging process can damage the signal beyond repair; it may have noise or some non-linear transformation applied to it that renders it useless.
 

wildbillx

Member
Premium Subscriber
Joined
Feb 25, 2006
Messages
400
Location
Cape May, NJ
#20
I have it working except it mutes the volume. Is there a screen shot of what the info is suppose to look like?
 
Last edited:
Top