Thanks for the transmission Stefan. although this is easy decode, it can
be used to optimise the receiver for longer messages.
I've tried to process this using different vtblank -a and -t parameters,
use ebnaut in signal analysis mode (knowing thr message in already) and
graph the resulting carried Eb/N0.
The pipeline is: vtread ... | vtfilter -h bp,f=6470.1,w=3000 | vtblank
-aXX -d 0 -tYY |vtmult -f 6470.1 | vtresample -r 240 | ebnaut -r240 -other
parameters -d15 -f16 -M received_message. Only the -a and -t parameters
are changed. 14 -a values and 409 -t values were used (5726 calculations,
about 1 day on 48 intel cpu cores).
The attached graph shows the carrier Eb/N0 for different parameters, and
can be used to optimise the receiving of longer messages. For this
transmission the sweet spot was -t 7500-8600 (for -a 1.7) and -t 8550-8750
(for -a 1.8), i'd use vtblank -a 1.7 -t 8050 -d0.
Beware that this is specific to qrm at this particular location, time of
day, thunderstorm activity etc. Beware2: this is for vlfrx-tools 0.7q
(which is unpublished bu Paul, but downloadable by tweaking the url, the
official 0.7p works similar), vtblank from version 0.8 behaves differently
(haven't checked why yet)
I can publish the scripts and the raw data in case anyone is interested.
VY 73
Jacek / SQ5BPF
On Mon, 5 Jun 2017, DK7FC wrote:
Date: Mon, 05 Jun 2017 14:41:28 +0200
From: DK7FC <[email protected]>
Reply-To: [email protected]
To: [email protected]
Subject: VLF: New EbNaut announcement...
Hi VLF,
For some local tests i like to transmit the 2 character message that i already
sent on 970 Hz 3 weeks ago. This time on 6470 Hz....
Such a short message with sch long symbols may be interesting for newcomers or
various tests by the established VLF receivers...
f = 6470.100000 Hz
Start time: 05.Jun.2017 13:00:00 UTC
Symbol period: 60 s
Characters: 2
CRC bits: 3
Coding 16K21A
Duration: 9h, 20min
Antenna current: ~ 460 mA
73, Stefan
blanktest_20170605.png
Description: PNG image
|
