DSD FME

[lwvmobile]

Is that a problem leaving it at 999?

This could be an issue, if the control channel rotates to another frequency, then you won't have a correct channel mapped in if call grants occur on that frequency if or when a call grant goes there, assuming that the control channel does rotate eventually. Some systems seemingly never rotate, others rotate frequently.

If it were me, I'd just load both 152 and 153 as 936900000 and later on, if you see 152 or 153 in a call grant, then you'll know which one was correct. Its entirely possible I was 'off by one' in the calculation, I never do the calculations, I was just trying my hand at it to see if I came up with a reasonable value. The issue is, for whatever reason, ETSI didn't see the value in having the control channel broadcast its own LCN value. There are a few places they could do so, but depending on if its CapMax, or Standard Tier 3, they don't really seem to ever be consistent in the fields you might expect to find them. I spend way too much time recently just poking around DMR CSBKs that carry broadcast parms and lpcn's looking for a consistent setup to get the control channel LCN, and it was never consistent, or just always the wrong value from the expected/known value.

 

[n5pwp]

git clone https://github.com/lwvmobile/dsd-fme.git
cd dsd-fme
git checkout audio_work
mkdir build
cd build
cmake ..
make -j `nproc`    <<<<<<<<<<
sudo make install

Hit a stumbling block. When I issue the command indicated above I get an error:
make: *** no rule to make target 'nproc'. stop.

Mike

 

[lwvmobile]

those are ticks ``` (the key to the left of the 1 key on the top row), not single quotes ''' , just run make if make -j `nproc` doesn't work.

 

[n5pwp]

One of the dangers of using a remote connection to another machine. I tried to build this on the machine that normally serves as the SDRpp server. I forgot that when you do a ctrl-alt-t it is issuing a terminal on the remote machine not on the one I'm wanting to build on. Once i caught myself I went back and removed all the stuff I had started. I then went back to the server machine and started over. Apparently since I had DSD-FME loaded on the laptop it found MBE. Now that I'm building it on the server machine it can't find it. Is that not carried over when I git clone? Now I'm getting an error "Could not find MBE: MBE_LIBRARY MBE_INCLUDE_DIR"
Hopefully I haven't hosed the good install on the laptop.
Mike

 

[lwvmobile]

No, nothing should be borked. If building fails along the way, no harm is done. Worst case scenario, you delete the contents of the build folder and start again at cmake .. after getting the dependencies installed.

If you are installing on another machine, you'll need to install all the dependencies first. Usually, the easiest way to accomplish this is just to run the script that automatically installs everything first, then you can install or checkout the audio_work branch and it should just build.

wget https://raw.githubusercontent.com/lwvmobile/dsd-fme/main/download-and-install.sh
chmod +x download-and-install.sh
./download-and-install.sh

Might be easier to just copy and paste things into the terminal so you don't make any transcription errors.

dsd-fme/examples/Install_Notes.md at main · lwvmobile/dsd-fme

Digital Speech Decoder - Florida Man Edition. Contribute to lwvmobile/dsd-fme development by creating an account on GitHub.

github.com

 

[n5pwp]

Well, after banging my head against the wall it appears I will have to take another approach to this. Apparently my i3 Phenom machine with 8GB of RAM is more underpowered than the i3 laptop with 4GB RAM. When I try to run SDRpp it nearly maxes the CPU out. I'm running about 93%, I will have to figure out how I can run both DSD's on the laptop. I did get your dev version running but the SDRpp used up so much CPU it drug the machine down. SDRpp is working fine on the laptop but since I removed the -i tcp since we're using the audio sink now it plays the audio rather than passing it to DSD-FME-dev. Hmmm, i went back to using the sdrpp server and set the input back to tcp and now it is back to working. Now i just need to work on getting your dev version running on the laptop. I seem to even be getting a stronger signal with the RSPDuo hooked directly to the laptop instead of going through the desktop machine. I don't understand that either. I need to research some more.

I'll let you know.
Mike

 

[lwvmobile]

How much bandwidth/sampling are you trying to run the RSPDuo at? The entire 10 Mhz? Or a lesser value? I'd go down to 1 or 2 Mhz if possible, maybe that will help with the CPU usage, no need for all the bandwidth if you just run a single trunking instance.

Apparently my i3 Phenom machine with 8GB of RAM is more underpowered than the i3 laptop with 4GB RAM.

So, is it a Core i3, or an AMD Phenom II X4? If its an i3, what generation? I have an older Core i3-5020U laptop with 6GB RAM and SDR++ and DSD-FME run perfectly fine on it without pegging it out. Heck, for that matter, I have an even lesser powered AMD E2-7110 APU with 4GB RAM and that doesn't get pegged out either. Again, that's just with an RTL dongle running at 1.024 Mhz Sampling.

 

[n5pwp]

How much bandwidth/sampling are you trying to run the RSPDuo at? The entire 10 Mhz? Or a lesser value? I'd go down to 1 or 2 Mhz if possible, maybe that will help with the CPU usage, no need for all the bandwidth if you just run a single trunking instance.

This is the config for the DX4300. I stepped the Sample Rate down to 2M from 5M and it only dropped the CPU usage 4%. Even without it running (the app is up but not receiving) it still maxes it out. I think I started to run SDRpp on the DX4300 before and it had the same issue. That's why I went to running the SDRpp server on the DX4300 and stream it to the Acer laptop. I guess it's the difference between a 13 yr old desktop machine and a 5 year old laptop.


The Acer laptop seems to run it just fine.

I am running X window system on the DX4300 and its Wayland on the laptop.
Also I noticed that SDRpp on the laptop has lost the Sample Rate dropdown on the laptop. I even loaded the latest nightly release but its just gone.

I need to go find where SDRpp is installed and see if there is a config file that controls that. I used to be able to run up to the full 10MHz on the laptop.
Mike

 

[lwvmobile]

Also I noticed that SDRpp on the laptop has lost the Sample Rate dropdown on the laptop. I even loaded the latest nightly release but its just gone.

Just doing a quick search of the SDR++ github, the sample rate may be available if you use the SoapySDR module, assuming its available/installed for you, and assuming what is displayed in this image is still valid. Either that, or you may need to do something to configure the device independently of SDR++, maybe in its own API. I really have no idea about SDRPlay devices since I don't have one.

I guess it's the difference between a 13 yr old desktop machine and a 5 year old laptop.

Yeah, that could quite possible be the difference.

EDIT: Another thing to try is GQRX, see if running with that lowers the CPU consumption any. With a laptop that old, I kind of wonder if what's hogging the CPU is Ubuntu running X in software rendering mode. No idea if that old radeon chip is still supported by mesa or not. It probably is, but worth a glance I suppose.

 

[n5pwp]

Just doing a quick search of the SDR++ github, the sample rate may be available if you use the SoapySDR module, assuming its available/installed for you, and assuming what is displayed in this image is still valid. Either that, or you may need to do something to configure the device independently of SDR++, maybe in its own API. I really have no idea about SDRPlay devices since I don't have one.

I'm using SDRPlays's API file. All you do is run the SDRPlay.run file and its up and running.

I'm on the SDRpp Discord. I will bring it up on there and see if anyone has any ideas.
Mike

 

[n5pwp]

EDIT: Another thing to try is GQRX, see if running with that lowers the CPU consumption any. With a laptop that old, I kind of wonder if what's hogging the CPU is Ubuntu running X in software rendering mode. No idea if that old radeon chip is still supported by mesa or not. It probably is, but worth a glance I suppose.

Actually the laptop is the newer machine. The DX4300 is 13 years old. The laptop is 'only' 5.

 

[n5pwp]

and then you'll have the latest bleeding edge version installed. What you can do if you want to switch between versions later on on is just open the build folder for main or audio_work in the terminal (browse to it and right click and open terminal here) and just run the command sudo make install and that will install the main branch version, or the audio_work version if you are in their respective build folder.

OK, got the Sample Rate worked out. I ran the download-and-install.sh script and then went to the build directory and ran sudo make install. I issued my same start command as before and it still launches the Main Build version. I still use dsd-fme, right?
Mike

 

[grosminet]

Need help with DMR capacity plus. As you may be noticed, we have world rugby cup in France . Some of our stadium increased their DMR capacities . I have an TRX-1 to follow the traffic but I want to follow with dsd-fme . my questions :
Under dsd fastlane I have " TLC tier III PC grant" ans cap+ restch=1, is it a tiers III or a capacity plus network ?
with dsd-fme, the capacity_plus_channel is asking for LSN , but I found LSN are changing , so I how can I be sure for the right LSN

thanks for answer et sorry English is bad I need some more practice

 

[n5pwp]

OK, got the Sample Rate worked out. I ran the download-and-install.sh script and then went to the build directory and ran sudo make install. I issued my same start command as before and it still launches the Main Build version. I still use dsd-fme, right?

Wow, this is really getting confusing. I created the dsd-fme-dev folder and ran the download-install script. I ran dsd-fme start command and it launches the main build of the app. I went back to the release folder and ran sudo make install. It successfully completed and when i ran with the regular start command it ran the AW build! I know for certain that I built the Audio branch in the dsd-fme-dev folder. How the heck did it get over into the main folder build?

Oh yeah. The TG hold seems to be working just fine. It actually displays it on the screen.

Mike
Update: I just looked and it appears that I used your wget commandline and it is pointing at the main library. When I ran that in the dsd-fme-dev directory it actually downloaded the main files. I should've changed main in the line to aw.

wget https://raw.githubusercontent.com/lwvmobile/dsd-fme/main/download-and-install.sh

 

[lwvmobile]

Wow, this is really getting confusing. I created the dsd-fme-dev folder and ran the download-install script. I ran dsd-fme start command and it launches the main build of the app. I went back to the release folder and ran sudo make install. It successfully completed and when i ran with the regular start command it ran the AW build! I know for certain that I built the Audio branch in the dsd-fme-dev folder. How the heck did it get over into the main folder build?

Usually, the simplest explanation is the correct one. You most likely ran git checkout audio_work in your main install folder, and not the one you set up. It honestly doesn't matter so much, you can just flip and flop back to either or if you know which one is which, or just run git checkout main and cmake and make again and it'll change to either which one you want.

Update: I just looked and it appears that I used your wget commandline and it is pointing at the main library. When I ran that in the dsd-fme-dev directory it actually downloaded the main files. I should've changed main in the line to aw.

As far as the scripts go, they all point to main, even on other branches, I don't update those for new branches, as the new branches are meant to usually either merge into main eventually, or just outright replace main eventually.

 

[n5pwp]

Thanks again for your help and your patience. wasn't sure if I should take this to PM but then I thought if anybody is in the same situation I am it might be helpful and you won't have to go through this again. the audio_works version seems to be working fine.

Mike

 

[lwvmobile]

Need help with DMR capacity plus. As you may be noticed, we have world rugby cup in France . Some of our stadium increased their DMR capacities . I have an TRX-1 to follow the traffic but I want to follow with dsd-fme . my questions :
Under dsd fastlane I have " TLC tier III PC grant" ans cap+ restch=1, is it a tiers III or a capacity plus network ?
with dsd-fme, the capacity_plus_channel is asking for LSN , but I found LSN are changing , so I how can I be sure for the right LSN

Not quite sure what fast lane displays on a cap+ system when you have a call grant, or channel activity in a slot, but it almost sounds like you have two frequencies that belong to different systems. Not entirely sure.

As far as needing to map those cap+ lsn frequencies, you need to look at the example cap+ csv file in the examples folder and fill it out with known frequencies. What I've typically found to be true on Cap+ systems is that when it goes into beacon mode (the short busts) that exact frequency is the rest channel at a given moment. Cap+ LSNs also operate in pairs, so LSN 1 and LSN 2 are the same frequency, LSN 3 and LSN 4 are the same frequency, so you'll need to load the same frequency for both.

Anyways, more information on loading up and trunking a Cap+ system can be read in the example usage, and also by checking out the trunking.sh file for more information.

dsd-fme/examples/Example_Usage.md at main · lwvmobile/dsd-fme

Digital Speech Decoder - Florida Man Edition. Contribute to lwvmobile/dsd-fme development by creating an account on GitHub.

github.com

dsd-fme/examples/trunking.sh at main · lwvmobile/dsd-fme

Digital Speech Decoder - Florida Man Edition. Contribute to lwvmobile/dsd-fme development by creating an account on GitHub.

github.com

 

[n5pwp]

One more observation. Everything working great. But, I did see that if I lower SR to 2MHz I receive no signal. If I raise it to 3 MHz it appears on the spectrum but it produces very choppy decodes. But, if I raise it to 4 MHz then it is all crystal clear. So it seems this site would not work well with an RTL-SDR dongle with only 2 MHz.
Mike

 

[lwvmobile]

One more observation. Everything working great. But, I did see that if I lower SR to 2MHz I receive no signal. If I raise it to 3 MHz it appears on the spectrum but it produces very choppy decodes.

How does NOAA Weather Radio sound when listened to at those rates?

 

[n5pwp]

How does NOAA Weather Radio sound when listened to at those rates?

At 2/4 MHz they sound the same. At 3 MHz there is no audio but I can see the signal on the spectum/waterfall.