Thursday, January 7, 2016

Levels and Deviation

One of the aims of the MMDVM is to ensure that setting it up should be easy. For example set up a repeater and then play traffic through it, set the levels so that the traffic is audible on the desired mode, and the levels for the other modes should also be correct. As specified each data mode has a fixed set of deviations associated with it. Therefore the relationships between each is fixed and that is how it will be in the modem. Getting these levels right is not easy though.
 
For example on my old IC-E92D I can have the D-Star being transmitted with a wide range of deviations and it still receives perfectly. The newer ID-51a are not so lenient, but I don’t have one of those. Last night I heard my first DMR through the MMDVM, it was only for a few seconds, but it was there J However the levels probably were borderline and I wasn’t able to reproduce it, I think that my CS700 has pretty tight filters. I suspect that level setting will become a bit of an issue in the future. I’d imagine a tool using something like an AirSpy or similar could come to the rescue ultimately for setting things up.
 
I’m still at the stage of experimenting with the relative levels within the MMDVM, and I hope that further on-air testing elsewhere will enable the correct relative levels to be set once and for all.
 
So what of the levels. D-Star uses GMSK with a 2.4 kHz overall deviation and that fits into a 12.5 kHz channel very nicely. DMR uses 3.888 kHz overall deviation and still fits into a 12.5 kHz channel, it has to, it’s a commercial standard and must operate within a much tougher regulatory framework than we amateurs. But what of System Fusion? The spec says it has an overall maximum deviation of 5.4 kHz, which does not fit into a 12.5 kHz channel. It may be fine for the American 15 kHz channels but not our bandplans. Putting two closely sited System Fusion repeaters on adjacent 12.5 kHz channels and expecting happiness is not on the cards. There was some talk of banning System Fusion in Germany because of this, but I don’t know what the outcome was. Yaesu said that the deviation couldn’t be reduced. Hmmmm, let’s see.
 
Despite a lot of obfuscation within the specification, the modulation method and constants for System Fusion turn out to be the same as for DMR. Obviously the content of the messages are complexly different, but it’s all C4FSK (C4FM is a Yaesu-ism) which is sent over the air. So if DMR can live within tighter limits then why can’t System Fusion? Only Yaesu can answer that.
 

No comments:

Post a Comment