- Status
rob12770
Member
he was askin about dmrdecode not dsd
BM82557
Member
That post I pointed him to has a zipped copy of DMRDecode. The DMRDecode website in the UK has been down for a while.
rob12770
Member
dmrdecode dosent seem to do anything anyway,,
The download site is github and is working fine. Ian has used the university server to host example images of the software running and it is this that doesn't seem to be available now.
Hi Eric
Sorry been offline for a bit
It seems they have wound the transmitter power back or something has gone up between my place and the site as the signal is now very noisy at home and I can't get reliable decodes at all.
It will mean a drive with laptop and should be able to do that later this week - lots of work commitments at the moment. They are not using the system very much so I'll need to pick a time when they are.
I use a USB SDR so I hope it will be good enough.
Cheers
Is http://borg.shef.ac.uk/dmrdecode/ still active? I keep getting a "The connection has timed out" error.
BM82557
Member
No, see http://forums.radioreference.com/trunking-control-channel-decoding/303215-dmrdecode-build-74-a.html for info (also includes a copy of the latest build for download)
I have a somewhat random question: Does DMRDecode log data from any frequency in a DMR system, or does it need a special freq such as a control channel-like frequency, to log the relevant data?
I will show data on voice or control channels.
very frustrated
So far I've not been able to decode a single DMR frame with DMRDecode. The most I've gotten is endless repeats of this:
10:40:34 PM DMR BS Voice Sync Acquired
10:40:34 PM jitter change to 6
10:40:34 PM jitter change to 3
10:40:34 PM jitter change to 5
10:40:34 PM jitter change to 2
10:40:34 PM Sync Lost : centre=1844 max=15289 min=-11600 umid=10247 lmid=-6558
I'm running Kali linux on a 3 Ghz AMD box with 4 Gigs of Ram. My source is a RTL SDR dongle connected to a Larsen tri-band antenna. I'm able to decode and hear voice frames from numerous DMR signals with DSD and GQRX, but DMRDecode won't recognize any of those same frequencies.
I've tried inverting the signal and messing with the audio gain in GQRX such that DMRDecode's audio level is a fairly steady green, all to no avail. When I run DMRDecode, the audio is piped from GQRX's UDP stream via PADSP, which works great with DSD. AFAIK I'm getting audio into DMRDecode based on the green and red audio level indicators. As I understand it, the audio output from GQRX's UDP stream is 48 Khz, so there shouldn't be any sampling rate issues. Here is my command to start DMRDecode:
"nc -l -u -p 7355 | padsp -d java -jar dmrdecode_b74.jar"
Obviously I'm doing something wrong, but I'm not sure what. It's possible that in my particular location I can't get a strong enough signal for DMRDecode to work, but that would mean DMRDecode needs a stronger signal than DSD. Is that possible?
As far as I can tell, the DMR signals I'm seeking to decode are very strong based on the spectrum graph in GQRX as well as listening to them in NFM mode on my hardware scanner. (Even though a hardware scanner can't decode DMR signals, you can still discern rough signal strength and obvious problems with SNR/interference just by listening to the audio in NFM mode). Obviously I'm missing something here, any ideas what it might be?
So far I've not been able to decode a single DMR frame with DMRDecode. The most I've gotten is endless repeats of this:
10:40:34 PM DMR BS Voice Sync Acquired
10:40:34 PM jitter change to 6
10:40:34 PM jitter change to 3
10:40:34 PM jitter change to 5
10:40:34 PM jitter change to 2
10:40:34 PM Sync Lost : centre=1844 max=15289 min=-11600 umid=10247 lmid=-6558
I'm running Kali linux on a 3 Ghz AMD box with 4 Gigs of Ram. My source is a RTL SDR dongle connected to a Larsen tri-band antenna. I'm able to decode and hear voice frames from numerous DMR signals with DSD and GQRX, but DMRDecode won't recognize any of those same frequencies.
I've tried inverting the signal and messing with the audio gain in GQRX such that DMRDecode's audio level is a fairly steady green, all to no avail. When I run DMRDecode, the audio is piped from GQRX's UDP stream via PADSP, which works great with DSD. AFAIK I'm getting audio into DMRDecode based on the green and red audio level indicators. As I understand it, the audio output from GQRX's UDP stream is 48 Khz, so there shouldn't be any sampling rate issues. Here is my command to start DMRDecode:
"nc -l -u -p 7355 | padsp -d java -jar dmrdecode_b74.jar"
Obviously I'm doing something wrong, but I'm not sure what. It's possible that in my particular location I can't get a strong enough signal for DMRDecode to work, but that would mean DMRDecode needs a stronger signal than DSD. Is that possible?
As far as I can tell, the DMR signals I'm seeking to decode are very strong based on the spectrum graph in GQRX as well as listening to them in NFM mode on my hardware scanner. (Even though a hardware scanner can't decode DMR signals, you can still discern rough signal strength and obvious problems with SNR/interference just by listening to the audio in NFM mode). Obviously I'm missing something here, any ideas what it might be?
Evidently Jeremym70 had this exact same problem with a disc tap setup, as described at post #543 circa 2/28/2013 (in this same thread). Apparently Jeremy solved his problem with a 10k resister. At this point I'm wondering if replacing my soundcard might help? Perhaps my soundcard is doing something funky with the baseband audio coming out of GQRX, is that a possibility? Any ideas on how to isolate/test my signal chain to narrow down the problem?
The DMR systems I'm monitoring originate from repeaters of varying signal strength and site distance, but several use huge mountaintop repeaters with paperclip reception. Has anyone else had any issues with DMRDecode and a RTL SDR 2838?
Have you tried using DSD+ at all? It'll give you all the same information as well as decode the audio. Plus it has a waveform viewer to see your source audio. Very handy for determining if your level is too hi/low/noisey etc. Might be worth a look.
Update: Forts, I have used DSD+ and it is really cool, but I also want to use DMRDecode in Linux with GRQX. I may have just found part of my problem. Although I was starting DMRDecode with the PADSP command, I just realized that Java programs don't recognize Pulse Audio (at least not without modifications). When I use SDR programs compatible with Pulse Audio (i.e. DSD, Multimon, etc) I never have to mess with audio input gain.
However, DMRDecode is very sensitive to small input gain changes from GRQX and/or pavu/soundcard audio controls. I ran "pacmd list-sinks" and it confirmed my soundcard was 48Khz, and I set the default sample rate in the Pulseaudio config file to 48 Khz. I guess I need to mess with the audio gain settings and see if that helps...
However, DMRDecode is very sensitive to small input gain changes from GRQX and/or pavu/soundcard audio controls. I ran "pacmd list-sinks" and it confirmed my soundcard was 48Khz, and I set the default sample rate in the Pulseaudio config file to 48 Khz. I guess I need to mess with the audio gain settings and see if that helps...
Well, I just broke down and re-formatted my computer as a dual-boot machine with Windows 7 & Kali Linux. DSD+ and SDR# are WAYYY better than DMRDecode/DSD/GQRX. I was able to get DSD+ working with vbcable and SDR# immediately on Windows 7 with zero problems. DSD+ is freakin amazing. It is really cool that DSD+ decodes the audio & logs the relevant data all within the same application, but in different screens that you can move around. [Note: I previously using DSD+ on my OS X machine running parallels, but it was a nightmare to configure and I never got audio to decode right. But, now it all works great in dual-boot mode]
I'm also shocked at how much better I like SDR# compared to GQRX; I had no idea SDR# allowed you to scroll through the spectrum graph in discrete steps linked to your step size setting, I sure wish GQRX had that feature. Come to think of it, I recall reading that GQRX is basically EOL for the indefinite future...
Last edited:
Further update: Well, DMRDecode works awesome on the "Windows 7 side" of my dual boot machine, with SDR# and VB-Cable. I was also surprised to learn I could run DMRDecode AND DSD+ simultaneously with VB-Cable. DMRDecode actually complements DSD+ quite nicely.
Obviously my inability to run DMRDecode in Kali Linux via GQRX had nothing to do with my soundcard, dongle signal quality or CPU Capacity. Perhaps the audio stream was somehow getting messed up.
Obviously my inability to run DMRDecode in Kali Linux via GQRX had nothing to do with my soundcard, dongle signal quality or CPU Capacity. Perhaps the audio stream was somehow getting messed up.
johnnyelectron
Member
- Joined
- Dec 19, 2002
- Messages
- 201
Can audio be decoded?
I'm just jumping in here, but is it possible to decode the actual MOTOTRBO digital DMR audio from (slot 1 or slot2) on a given system, or only the control channel data using a SDR radio?
I'm just jumping in here, but is it possible to decode the actual MOTOTRBO digital DMR audio from (slot 1 or slot2) on a given system, or only the control channel data using a SDR radio?
DMRDecode does not decode audio. You need DSD or DSDPlus for that.
Andru, Forts mentioned in exactly the post before yours this software does not decode voice. It only prints control channel messages. You need DSD+ or WinDSD to decode voice.
- Status