ssb mode with tiny carrier....
bbruhns at erols.com
Mon Jan 7 18:03:02 CST 2008
ACSSB - Amplitude Compandored Single SideBand.
2:1 compansion, like dbx, which was rather effective at noise reduction - and, that bit of carrier in the middle of the passband,
to use as a level and frequency reference, made things a bit easier on the receive side - but the receiver needed the extra
circuitry to grab onto that little bit of carrier.
There are three methods of SSB generation: filter, 90 degree audio-rf phasing, and a sort of phasing method known as Weaver, which
is I-Q with a hypothetical 'carrier' in the middle of the sideband. The Weaver method could simply have been slightly deliberately
misadjusted in the carrier balance to generate the ACSSB reference carrier, and this made it perfect for ACSSB generation.
I guess the problem with ACSSB was that generating and receiving it was more complex than FM, and frequency control and stability
was more critical. If two stations transmitted at the same time, frequency error would make some odd sounds. And then it was a
completely different technology, requiring linear amplifiers, etc, which made for an expensive start-up. Meanwhile, the UHF bands
were opening up with advanced cellular and trunking technology, and most users preferred to stay on FM and go digital. If we had to
go with linear amp stages, we might as well go with poly-point M-ary, and go digital. And here we are today.
Hams could use ACSSB, but there will be some disputes about the proper amount of inserted reference, which can make some awful loud
or low audio levels if misadjusted. I think that it would be better in 2008 to run ordinary SSB with normal peak-AGC, and add DSP
auto-clarifying AFC on the receive side. I saw an article about auto tuning of SSB receive frequency recently, and it looks like
this technology has arrived. As for the 2:1 compansion, the analog compander ICs seemed to be getting hard to get a few years ago,
but if we have a DSP to do frequency adjustments, we -could- also use it for 2:1 dynamic compansion.
The fast 2:1 compansion (if used) would be in addition to the normal relatively slow receive AGC, and it would reduce perceived
background noise. But in local-distant VHF, we would also want to have some kind of dynamic decay time - slow while receiveing a
signal, to avoid lots of noise between syllables, but fast without a signal, to jump to high gain quickly, to hear weak replies to
strong signals. That inserted reference carrier might be used as a signal-present reference to do that, but I think we should
ignore it as a level reference, to avoid fistfights among hams. It isn't really needed. If we do that, we would be compatible with
ordinary SSB, although they would hear a whistle... which their DSP would suppress... and they would hear our extra transmit
compression, which would make our acoustic background noise louder to them (no biggie).
If we use DSP, we could apply some more noise reduction on the receive side with that, in addition to the 2:1 compansion. This
could be dynamically applied to minimize loss of quality under strong-signal conditions.
Maybe the reference carrier could tell us that the signal was also using 2:1 compansion. So the receiver would know whether it was
hearing an encoded signal, and it could kick in the expander.
----- Original Message -----
From: "Mike O'Dell" <mo at ccr.org>
To: <tacos at amrad.org>
Sent: Monday, January 07, 2008 5:52 PM
Subject: ssb mode with tiny carrier....
> once upon a time we had a speaker talk about an SSB mode
> which, i believe, compandored the audio, but also
> left a little bitty stump of carrier in the signal
> so receivers could autotune and AFC. now i can't
> remember the name of it - ACSB ??? wazzat right?
> my question is "why didn't that ever get explored
> on the ham bands?" with the advent of SDR, it would
> seem to be a lot easier to examine such things
> in a "live fire" setting.
> what am i missing here??
> Tacos mailing list
> Tacos at amrad.org
More information about the Tacos