Stefan wrote:

> f = 6470.00000 Hz
> Start time: 16.05.2016   21:00:02 UTC
> Symbol length: 100 s
> Characters: 20
> CRC 16
> Coding 8K19A

I wasn't able to decode the 18 char message.  Thanks to
Markus for the sneaky decode, I was able to reverse the
modulation to see the signal was 13.6dB in 8.8 uHz.

That puts Eb/N0 at about -6.5 dB.

Strange, I was expecting a better result because the
background noise was quite low throughout the message.

Tomorrow I'll have a closer look to try and see if
propagation is upsetting the phase.

--
Paul Nicholson
--