[RTTY] New proposal for a Automated Digital Contest

[Terry]

Hi Chen,

That's great technical info.  

These kinds of inputs are great as they cause us to think along other lines.
For example, one idea might be to do a 2Tone/MMTTY shootout where every QSO
is make once with 2Tone and once with MMTTY.   That experiment would show
practical results that could be compared with the theoretical side of
things.   Who knows, we may find that on certain bands, that MMTTY is better
than 2Tone.

I started a C# program as a proof of concept.  I got it launching 2Tone and
receiving RTTY data.   I just added a timer where it will calls CQ ever 15
seconds so it's now transmitting.  I'm thinking for the first pass it will
like a simple remote BBS.   Something like if you answer the CQ as AB5K de
W7AY W7AY (call twice) then it will respond with something like W7AY de AB5K
UR 599 in TX 73 de AB5K.   

The contesting version would be machine to machine and we could implement a
checksum for data validation along with a "AGN?" response if the data does
not validate.  The number of repeats might be part of the logbook entry as
that's valuable for propagation studies.


Terry  AB5K



    



-----Original Message-----
From: Kok Chen [mailto:chen at mac.com] 
Sent: Saturday, April 19, 2014 3:15 PM
To: RTTY Reflector
Cc: Terry
Subject: Re: [RTTY] New proposal for a Automated Digital Contest


On Apr 19, 2014, at 12:35 PM, Terry wrote:

> There are all kind of interesting things that we could experiment with.
I
> believe that MMTTY already supports 850 HZ shift.  If RM-11708 is adopted,
> we might find a 2.8 KHz waveform that we want to experiment with.   From
> what I read on the Rm-11708 FCC comments, conserving bandwidth is 
> outdated and it's all about efficiency,  so we need to get out of the
"old-fashioned"
> thinking that conserving bandwidth is the right thing to do.

With modern modems like 2Tone and fldigi, the demodulators are implemented
as separate OOK demodulators at the Mark and Space frequencies.  So you
don't have any disadvantage by going to a wider shift.  Each tone has its
own narrow ISI-free data filter. (I can't speak for MMTTY, since I could not
find documentation on whether the filters used by MMTTY are ISI-free, nor
documentation on any selective fading countermeasures.)

*However,* you also gain very little by widening the shift.  

850 Hz shift will help marginally in terms of selective fading.  170 Hz
shift is at about the cusp of being able to take advantage of frequency
diversity when there is HF selective fading.  You will be able to notice
that 850 Hz works better at times, but probably not worth eating up that
much more bandwidth.  Most of the time, you will see no practical
difference.

Take a look at your crossed ellipse display.  Do you see the amplitudes of
the mark and space ovals fluctuate independently?  If you do, 170 Hz is wide
enough for modern demodulators to take advantage of this "frequency
diversity." I.e., when Mark fades, the demodulator shifts to more of a
"Space-only"  demodulator.  

It is only when both 170 Hz Mark and Space tones fade in and out
simultaneously ("flat fading") that you might be at a disadvantage when
compared to a 850 Hz shift.

If MMTTY does not have a good ATC implementation, then 850 Hz shift will do
you no good whatsoever over 170 Hz shift.

73
Chen, W7AY