Message-ID: <59E86EF4.1080502@posteo.de>
Date: Thu, 19 Oct 2017 11:23:00 +0200
From: DK7FC <selberdenken@posteo.de>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
References: <CAGZ2Mybw5K-2Tgu8Nyc0NqtN4JFgMqq=Eu0fsrSvKyMeg=HGVA@mail.gmail.com>
In-Reply-To: <CAGZ2Mybw5K-2Tgu8Nyc0NqtN4JFgMqq=Eu0fsrSvKyMeg=HGVA@mail.gmail.com>
Subject: Re: LF: 6470.1 Hz 16-char decode nr Moscow
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

Hello Alex, Paul, Markus and the rest of the small VLF group,

Indeed, a great achievement, one of the things i will look back to, when 
my live comes to an end some day.

Thanks to Alex/RN3AUS for building up his VLF equipment in the garden 
near Moscow, and for going there to take the daily files! Exemplary!!!
Thanks to Paul for his work, here the EbNaut software/decoder which 
works perfectly on VLF. I no other mode we would have reached that result!!
Thanks to Markus/DF6NM for writing/providing useful tools for Windows 
operators, here the add.exe program which allowed a weighted stacking of 
the files.
But we must not forget to think about Wolf/DL4YHF for writing his great 
SpecLab software, which is used and has been used by the most VLF 
operators since the beginning! SpecLab was used here on the tranmitter 
site to generate a stable frequency and phase, even correcting phase 
changes of the antenna system (caused e.g. by rain or by temperature 
changes of the coil or by soundcard glitches or or or). SpecLab has also 
been used to get the necessary accuracy of the RX frequency in the 
countryside as well as perfect timing.

We were quite lucky! Alex told me he will go to the graden to collect 
the daily txt files on monday, thus i decided to run a 11th 
transmission. Actually 'only' 10 transmissions were planned. No chance 
of a decode with just 10 days.
Patience is essential and it is another good example to never give up 
trying to do the 'impossible' things!
A plain carrier would have been easier because one can see something 
building up after a much shorter time. But a message which is suddenly 
decoded is much more fascinating :-) I lisented the song several times 
last evening :-)

The files were scaled/normalised by the square of the amplitudes, as 
suggested by Markus some time ago, i.e. kn[dB]= 2*20*log(U1/Un), where 
U1 is the amplitude of the first recorded day.
I tried linear scaling of the files and got a decode too, so the 
difference is very small.
Without daily amplitude scaling there was no decode.
In one day (10th), the QRN was very high, so the amplitude was scaled 
down quite much (-4.4 dB). But even when leaving out this worst day, 
there was no decode.

The set of data is very well suited to play with the add.exe tool by 
Markus. Unfortunately i have no Linux computer which had enough RAM to 
try to get results with Linux. Otherwise a decode may have happened 
earlier, when using the -a parameter.
The last 5 days were quite quiet. Maybe a 12th transmission on a quiet 
night would allow to get the decode after 'just' 6...7 days. But this 
information is unknown as long as there is no decode of course.

The 6470 Hz experiments are no completed here for some time. The grabber 
windows have been changed to 5170 Hz already. Last night i added more 
turns on the coil. Now there are 4 buckets fully wound with 0.25 mm 
enameled Cu wire. I checked the resonance frequency, i can tune down to 
5364 Hz now. About 140 mH missing. Today i will buy another bucket and 
start to wind a 5th bucket. A first carrier will soon appear at 5170 Hz. 
The show must go on :-)

73, Stefan


Am 18.10.2017 18:25, schrieb Alex K:
> Hello VLF!
> I am glad to report that the 16-char message of Stefan on 6470.1 is 
> successfully decoded nr Moscow. The stacking of 11 copies of 
> transmission with accurate weight summation was required. Stefan the 
> first has achieved successful decode from my files. Then with his 
> councils it was succeeded to make it also to me. My big 
> congratulations to Stefan! Excellent work and is a little good luck 
> :-). Also huge gratitude to Markus for the excellent utility add.exe 
> and of course to Paul for magnificent ebnaut!
> 73 de rn3aus/Alex
>
> PS all files are here 
> http://rn3aus.narod.ru/ebnaut-16ch-all-files-decoded.zip
>
> ebnaut-rx V0.7a
> input file C:\EbNaut\16char\sum-all-w.wav
> sample rate 0.342935 per second
> rx frequency 6470.1
> file start time 2017-10-15 15:43:32.290
> polynomial 16K21A
> crc size 19
> number of chars 16
> block size 115
> symbol period 20.000
> number of symbols 2160
> list length 868865
> start offset 998.574
> freq offset 0.000000
> cores 1
> skipped 997.273536 seconds to start
> initial reference phase -37.2
> ---------------------------------------------
> found YELLOW SUBMARINE
> list rank 746119
> reference phase  180  180  180  180
> carrier S/N 19.25 dB in 23.1 uHz, -27.11 dB in 1Hz, -61.09 dB in 2.5kHz
> carrier Es/N0 -14.10 dB
> carrier Eb/N0 -0.35 dB
> info bit period 473.68 seconds
> symbol error rate 852/2160 = 39.444 %
> Es/N0 from symbol errors -14.4 dB
> Eb/N0 from symbol errors -0.7 dB
> Shannon capacity 9.9 bits/hour
> Shannon efficiency 76.7 %
> symbols file C:\EbNaut\16char\sum-all-w-symbols.csv
> elapsed 126 seconds
> 1