Hi Paul,

Again thanks for the summary.

Does the message decode when correcting the phase? I think not.

73, Stefan


Am 22.12.2017 17:16, schrieb Paul Nicholson:

Stefan wrote:

> f = 17470.1000 Hz
> Start time: 18.Dec.2017  21:30:00 UTC
> Symbol period: 45 s
> Characters: 2
> CRC bits: 14
> Coding 16K21A

> Can you give me a feedback from the last night?

Now with 4 nights:

 18/19th Eb/N0 = -4.9 dB, S/N -39.6 dB in 1Hz, phase 110.4
 19/20th Eb/N0 = -6.5 dB, S/N -41.1 dB in 1Hz, phase 125.6
 20/21st Eb/N0 = -7.8 dB, S/N -42.4 dB in 1Hz, phase -22.5
 21/22nd Eb/N0 = -9.4 dB, S/N -44.0 dB in 1Hz, phase  55.6

No improvement on the earlier stacking: best is still
18/19 + 19/20 to give Eb/N0 = -2.7 dB

This isn't going to work without a repeatable night phase.

--
Paul Nicholson
--