I built an XLX941 DMR and YSF reflector on a Debian cloud computer. How can I permanently link the XLX reflector to my ASL3 hub? The ASL3 hub can be built on a Debian Linode cloud computer or a Debian physical computer. Is it possible to install both the XLX reflector and the ASL3 hub, permanently linked together on the same Debian Linode cloud computer. Or, are two separate Debian computers required for a XLX DMR and YSF reflector and an ASL3 hub?
AllStarLink is analog FM. XLXD is a hub for a variety of digital modes - notably D-STAR, YSF, and DMR. You cannot interconnect any of these modes without transcoding, such as an AMBE decoder.
You can build a sufficiently large Debian 12 server and install both Asterisk ASL3 and xlxd on the same system. But you’ll still need to do all of the work to setup transcoding between the modes. It is not a trivial configuration.
You will have to listen to the implementation of DVSwitch server on ASL3. Then, you can connect DVS to ASL I can do that with one type of Digital mode to ASL with a clearnode (HV).
You can use DVSWITCH to convert FM to DMR using md380-emu then link them together using MMDVM_Bridge. Below are from old notes. Join the DVSwitch group.io for more help.
[DVSWITCH]
sudo -s
cd /tmp
wget http://dvswitch.org/buster
chmod +x buster
./buster
apt update -y
apt install dvswitch-server -y
cd /opt
configure files
nano rpt.conf comment "; rxchannel = dahdi/pseudo ; No radio (hub)"
nano rpt.conf uncomment "rxchannel = USRP/127.0.0.1:34001:32001; GNU Radio interface USRP"
nano modules.conf uncomment "load => chan_usrp.so ;GNU Radio interface USRP Channel Driver"
Transcoding with ambe is only required for DSTAR. you do not need transcoding for DMR and YSF
On the same note transcoding from DSTAR to DMR to FM sounds really really bad. We have tried it many times with no real luck 
you can add Dvswitch server to ASL3 and run it in DMR mode direct to your xlxd server and it will be connected with no problems. I run it that want and have not had an issue. Please go to groups.io and go the the DVSwitch page under allstarlink for details.
then you are not using the 2 AMBE sticks to properly transcode.
This suggests there’s something wrong with your architecture. When building transcoders for multiple modes, you have to avoid transcoding any more times than necessary, because the DV vocoders are very lossy and the audio will degrade at each transcoding step. In fact the way you wrote your comment, if that was to be interpreted as a processing sequence is an example of bad architecture - D-STAR transcoded to DMR then transcoded to PCM or g.711u (for AllStar) won’t sound great.
This is one reason I won’t use transcoding XLX reflectors in multimode networks, the design of xlxd doesn’t allow access to the PCM stream in between the two vocoders, which would be the proper spot to take a signal out for ASL and other systems used for FM links.
I run a number of multimode transcoders that utilise a mix of DVSwitch, AMBE sticks and open source tools to bridge AllStar with DMR, YSF, M17, D-STAR, NXDN and P25, and achieve good audio quality, because there’s only one encode/decode cycle of transcoding between any two points of the network at most (0 between YSF, DMR and NXDN, which are simply translated, as they use the same vocoder). AllStar effectively only sees one vocoder between itself and any of the DV modes. On my systems, D-STAR actually sounds quite good when monitored on AllStar.
1 Like