Hello Markus,
I took your wav file that you provided for post analysis. Using the
amplitude normalisation function of the EbNaut Linux decoder i can get
a good Eb/N0 with a constant reference phase, see attachment.
The best result appears with -a80, just the symbol length. In my recent
970 Hz experiment i also got the best result with -a equal to the
symbol length.
73, Stefan
Am 12.08.2017 23:08, schrieb Markus Vester:
It
looks like we got a valid decode of Stefan's 16 hour transmission from
the DL0AO cardioid antenna.
I had also attempted an early decode from a file ending 18:12 UT. This
one appeared to be even better (rank 0, +2.2 dB carrier EbN0), perhaps
because it ended before the QRN increased.
We had no success with the earlier 24 h transmissions on Aug 10th and
11th, neither single days nor both stacked - surely due to much higher
QRN levels than today.
BTW Each decode attempt also produced several false hits with
relatively good ranking, so a longer CRC might be preferable.
All the best,
Markus (DF6NM)
|
YES.png
Description: PNG image
|
