> On Oct 24, 2016, at 1:42 PM, Jeff AC0C <keepwalking188@ac0c.com> wrote:
>
> I believe that Chen and David say DOOK is the best.
"DOOK" (differential on-off keying) is what David calls the method that 2Tone
uses to generate an RTTY signal. fldigi uses a similar scheme, but did not
give it any name (it is just "AFSK" to fldigi).
Basically, just like with good RTTY DEmodulators, Mark and Space are treated in
transmission as independent signals. As a result, each signal can be
individually wavehaped to very narrow modulated carriers (OOK).
Alex VE3NEA wrote a nice article on CW waveshaping in May/June 2006 QEX that
can be read by anyone with no DSP background. Imagine "DOOK" to be two
separate waveshaping of the Mark and Space carriers.
What 2Tone uses is not really "differential" in the sense of differential
paired signals (i.e., one is not the negative of the other). "Complementary
OOK" could be a better term, since there are parts of the waveshaped Mark that
overlaps with the waveshaped Space.
What is so bad about overlap in power? -- well, transmitter PA IMD, of course.
If you waveshape a lot to reduce the bandwidth, then the overlap will be
greater and the IMD will be greater.
You can actually see this in real life on a panadapter with signals that are
more than 40 dB about the noise floor. A 2Tone and fldigi signal will start
out looking like a part of narrow carriers.
They are very distinct in that the "hole" in between the Mark and Space
carriers are very deep. However, about 25 dB to 35 dB below the two carriers,
the signal will suddenly widen to 3 times the width at the peaks. Now, this is
still much narrower than an FSK signal.
When discussing the effect with K0SM, I had called this widened baseline a
"pedestal" or "mesa."
Andy K0SM and I had lots of fun at one time identifying the signals on the air
that shows this :-).
Rigs with poorer transmit IMD (like a K3) will have the pedestal appearing
earlier (around 25 dB), and better exciters when coupled with low IMD
amplifier, will have the pedestal grow from a lower level. One of the cleanest
signal I have seen came from a Yaesu rig running in Class A and into a good
amplifier.
I have yet to see an HPSDR on the air running predistortion (they call it "Pure
Signal") using these individually waveshaped Mark/Space carriers. When well
trained, a "Pure Signal" exciter can have IMD well below -50 dBc. That is the
future of amateur radio.
There is a lot of "art" involved in choosing how much, and what function of,
waveshape, to use to get a good balance of how narrow you want the signal, and
how much the temporal overlap causes IMD. And also not to waveshape too
heavily as to degrade demodulation on the receiving end.
I personally know that David pays *lots* of attention to this aspect. So, you
can be sure that when you run 2Tone under AFSK, you have one of the better
signals on the air (sounds sweeter to the ears too).
73
Chen, W7AY
_______________________________________________
RTTY mailing list
RTTY@contesting.com
http://lists.contesting.com/mailman/listinfo/rtty
|