Re: [Svxlink-devel] Voter + Siglevdetector
Brought to you by:
sm0svx
From: Tobias B. <sm...@us...> - 2009-09-29 22:31:35
|
Hi Adi, Thanks for the good bug reports. On Tuesday 22 September 2009 20:17:43 Adi Bier wrote: > Hi Tobias, > > I've played a bit with the voter configuration and found out a problem > that occurs sometimes. My configuration (at the moment) is a > svxlink-base and one remotetrx. The remotetrx is the relay-station with > RX and TX and the base has only a RX (I have my reasons why I have done > it in this way ;-). I've running the version 1681. > Here are some log entries: > > remote-TRX: > /Tue Sep 22 19:14:31 2009: Voter: The squelch is OPEN (QuantarRx=107.506) > Tue Sep 22 19:14:36 2009: Voter: The squelch is CLOSED (QuantarRx=4.09622) > Tue Sep 22 19:14:38 2009: Voter: The squelch is OPEN (QuantarRx=74.4778) > Tue Sep 22 19:14:51 2009: Voter: The squelch is CLOSED (QuantarRx=-4.20453) > Tue Sep 22 19:14:53 2009: Voter: The squelch is OPEN (QuantarRx=97.0731) > Tue Sep 22 19:14:53 2009: Voter: Mute(true) > Tue Sep 22 19:15:20 2009: Voter: Mute(false) > Tue Sep 22 19:15:22 2009: Voter: The squelch is OPEN (QuantarRx=75.9478) > Tue Sep 22 19:15:46 2009: QuantarTx: Turning the transmitter OFF/ > > The last two lines show that a strong signal (red lines) is being > received but it isn't detect by the svxlink base (blue lines). I happens > often, if a strong signal is received by the base first and a weaker > signal later by the remotetrx. > > svxlink-base:/ > Tue Sep 22 19:14:38 2009: Voter: The squelch is OPEN > (QuantarRx=73.2107) > Tue Sep 22 19:14:52 2009: Voter::satSquelchOpen(FALSE, QuantarRx): > Signal Strength = -4.20453 > Tue Sep 22 19:14:52 2009: Voter: The squelch is CLOSED > (QuantarRx=-4.20453) > Tue Sep 22 19:14:53 2009: Voter::satSquelchOpen(TRUE, OeblesRx1): Signal > Strength = 98.9846 > Tue Sep 22 19:14:53 2009: Voter::satSquelchOpen(TRUE, QuantarRx): Signal > Strength = 90.1575 > Tue Sep 22 19:14:53 2009: > Voter::chooseBestRx > Tue Sep 22 19:14:53 2009: Voter::satSquelchOpen(FALSE, QuantarRx): > Signal Strength = 0 > Tue Sep 22 19:14:53 2009: Voter: The squelch is OPEN > (OeblesRx1=98.9846) > Tue Sep 22 19:15:20 2009: Voter::satSquelchOpen(FALSE, OeblesRx1): > Signal Strength = -0.0725002 > Tue Sep 22 19:15:20 2009: Voter: The squelch is CLOSED > (OeblesRx1=-0.0725002) > Tue Sep 22 19:15:45 2009: Deactivating all links to/from > "RepeaterLogic" > Tue Sep 22 19:15:46 2009: QuantarTx: The transmitter is > OFF / > > If you need more information, please let me know. I'll have a look at it when I find the time. I created a ticket for it: https://sourceforge.net/apps/trac/svxlink/ticket/5 > Another point are the signal values that are logged in the files, e.g. : > / > Tue Sep 22 19:30:54 2009: Voter: The squelch is OPEN (QuantarRx=0.819237) > Tue Sep 22 19:30:54 2009: Voter: The squelch is CLOSED (QuantarRx=0.819237) > Tue Sep 22 19:31:38 2009: Voter: The squelch is OPEN (QuantarRx=-7.54016) > Tue Sep 22 19:31:38 2009: Voter: The squelch is CLOSED > (QuantarRx=-7.54016)/ > > I have configured SIGLEV_OPEN_THRESH=30 and SIGLEV_CLOSE_THRESH=10 and > of course conducted the siglevdetcal routine. > I think that the values in the logfile are the values _after_ the > SQL-delay-timers had been expired, the squelch is open and the signal > has been calculated/read out again and not the real values (above the > siglev_open_thresh) that were the reason to make the squelch going open. > So short bursts or unstable signals may trigger the squelch and the > sysop is wondering why. It's a minor "problem", do you see chances to > change it in further releases? Hmmm... that's actually the way it's supposed to work even if it looks a bit strange. The squelch delay was added to allow some time after squelch open has been signalled (serial squelch) until the signal level is measured. This is to allow the signal to stabilize a bit to get better measurements. In the siglev squelch case this behaviour may seem a bit strange though, I admit. What SQL_DELAY are you using? Does it help to increase it just a little bit? 73's de SM0SVX / Tobias > > 73's de Adi, DL1HRC > |