OK, Paul=2E
Stefan, it was neither GI nor= Ni=2E
you get some more opportunity:
carrier is on=2E
message= :
start 2016-02-17 18:00:00
same settings but::
repeat on hour= until the wx gets too bad=2E

Uwe/dj8wx

Von: Paul Nicholson <vlf0403@abe= lian=2Eorg>
Gesendet: 17=2E02=2E2016 10:59
An: <rsgb_lf_group@blacksheep=2E= org>
= Betreff: Re: AW: AW: Re: AW: Re: VLF: 8270=2E000 Hz and forest VLF = grabber preparations=2E=2E=2E

Uwe wrote:

>= ; message follows at
> 2016-02-1700 02:30:00
> same setting= s

No problem, decoded with Eb/N0 =3D +6=2E7, a bit weaker
tha= nks to the nighttime noise but would still do 10
second symbols=2E
Stefan wrote:

> Some decodes appeared but with 2 charac= ters it could
> be wrong=2E

These are false decodes, just = random noise which happens
to have a correct CRC=2E The default list= length of 20000
is too long for 2 chars, probably 1000 is enough=2E=

I'm still working on the difficult problem of finding the
op= timum list size and CRC strength for each message length=2E

As y= ou use longer list length, the chance of a decode increases
(with ra= pidly diminishing benefit) but the chance of a false
decode increase= s (roughly linearly)=2E A false decode is
more serious than a failed= decode because there's no way to
know if 'GT' or 'NI' was the corre= ct message! At least with
no decode, you know where you stand=2E
=
So to make use of longer lists, it is necessary to send with
str= onger CRC, but the extra CRC bits is more inner code overhead
which = reduces the chance of a decode (for a fixed signal
strength and dura= tion)=2E If CRC and list size are both
increased by the right amount= s, the decode chance overall
increases, but only up to a limit corre= sponding to around
3dB coding gain from the outer code=2E

The= best settings could be determined by a lot of brute-force
trials bu= t the permutations to be tested are many and the
computing time woul= d be huge, so I try to reduce the number
of trials by some performan= ce calculations=2E

For now, stay with 16 bit CRC and use list le= ngth 1000
or less for 2 chars=2E
--
Paul Nicholson
--
&n= bsp;