What is the P in IHOP?
My take on this:
Consider simple serial communication like RS-232- send a character with 8
bits, and an extra parity check bit. If the parity is incorrect, then
discard this character and ask for a retry. Better, send each character 3
times and accept it if two out of three match. Not foolproof enough? OK,
send each character 10 times and accept it only if they all match. Now it's
quite sure that was the character. But that is taking a lot longer. Send
these faster then. OK, but that requires more bandwidth so will need a bit
stronger signal in that wider bandwidth channel.
Go on to encoding 401. Add enough extra bits to the message to enable it to
be decoded correctly with a very very high confidence even if some bits are
missed here and there. How many extra bits this takes depends upon how many
characters are in the message. If the number of bits is a fixed maximum
length, that helps. If some messages are constrained to be specific formats
(for example, a space always follows a CQ, or a grid square is 4 characters)
that helps too since it reduces the number of different error combinations
to deal with. Trade these kinds of things off against the time used to send
the message, what type of modulation, and syncing, and the bandwidth, with
the objective being ability to decode very very weak signals in a chosen
band condition (e.g. weak E's). Obviously, the sender and the receiver need
to be using the same encoding and decoding.
Us grateful users have now identified other things we need or would like in
FT-8. To accommodate these in WSJT-X is likely not a simple tweak- the
structure details may have to change. So we need to be patient as this is
thought through.
I believe the "NA contest mode" was a tweak to try to make the process be a
bit more sensitive and efficient. Perhaps an unintended consequence was the
angst from mismatches of those in the mode and not. Plus with automatic
sequencing we may get lulled into thinking the computer is doing everything,
and forget that the realities of signals fading out from one sequence to the
next, or some QRM, or static burst are still there. If it was CW or SSB,
we'd try again to get the QSO, ask for repeats, turn the antenna, or choose
when to give up. It's all part of the game. New tactics need to be learned.
If the person answering your contest CQ on SSB comes back to you with "hi,
good to work you, my name is Fred and I live in Podunk Hollow." you would
usually patiently request his grid square and then end the QSO as quickly as
you could. Not the same on FT-8?
73,
Chet, N8RA
_______________________________________________
VHFcontesting mailing list
VHFcontesting@contesting.com
http://lists.contesting.com/mailman/listinfo/vhfcontesting
|