[Moon] wsjt-cw

[Klaus DJ5HG]

Hello Bernd, hello all,


just once more a comment here:


> Hello All,
> 
> Klaus writes of a "difficulty" of receiving the "address information" for
> the predefined set of callsign combinations and says it equals the
> difficulty to receive two arbitraty characters.
> 
> Actually, he does not mean "difficulty" as we usually understand it, he
> means the 'amount of information received'. This 'amount of information
> received' indeed is roughly equal in both cases. Only, this has nothing to
> do with how "difficult" it was to receive the signal over the air or whether
> only parts of the callsign were copied. It applies at FULL copy of all
> transfered dits and dahs, of all bits over the air.

Yes of course, even if we have a large SNR, then the DS 
decoder cannot receive more than about 14 bits of information. 
Even if the transmission is repeated, it will not increase 
this number in total. Even if the human operator could himself 
read the full callsigns by ear, the DS decoder by principle 
cannot decode more than these bits which are used to address 
the stored message in the list. That's the main point. The 
DS decoder cannot get callsigns at all, independently of SNR.

What I mean by "difficulty" is that it is easier to 
communicate 14 bits compared to 72 bits using the same 
transmitted energy at marginals SNR levels. So if the 
KV decoder of JT65 fails because 72 are too much then 
the DS decoder may succeed because it only needs to 
receive the address information of 14 bits.

All this is not different from other modes like SSB or 
CW. It is easier to get a report correctly than a 
report plus a contest number. The information will 
not increase with growing SNR, but the confidence will.
 
> BTW, in a sked (CW, JT65DS, SSB, whatsoever) the "difficulty" (as Klaus
> describes it) to receive the callsign data etc is way less than the
> difficulty to receive a single arbitrary character. Why? Because all
> information is already present at either side, be on hard disk or in the
> grey cells between the ears.

Yes that's a known fact. If I know all details of a signal 
then I cannot extract new communicated information from it. 
That was my basic question:  Is it the communication aspect 
which is important in a QSO or is it the simple physical 
detection of a signal? If the community only demands for 
the physical detection then I cannot find any serious 
argument why there should be reports and acknowledgements 
via the radio path. This would clearly be communication, 
not simple physical signaling. 

If the community demands for the callsigns to be 
COMMUNICATED via the radio path, then a sked with simple 
detection of a sufficiently fitting signal is not a valid 
contact. If the callsigns can be retrieved from the signals 
as if they would have been unknown (by the RS- or KV-
decoder in JT65 for example) then all is ok with the sked.

> 
> vy 73 Bernd DF2ZC (JO30RN)
> www.df2zc.de
> 


So as a result, I feel we both, Bernd and me, only differ 
in the rating of a QSO in respect to the views I mentioned 
above: The physical signaling vs. the communication aspect. 
This obviously is a question that waits for an answer from 
some higher level (preferably IARU). We here on the reflector 
will not solve it. I believe, we are at a stage where the 
main points and different views are clarified.


73,

Klaus, DJ5HG