Hello Marco,
Yes, you need some RAM to use the decoder, depending on the coding and
message length. 31 characters (in that given example) is quite long.
There may be shorter messages, for example just a callsign, that your
old PC may handle. Just try to use the same file and reduce the number
of characters to 20 or 10 or 6 so may find a maximum mumber of
characters that you can decode in 8K19A (the most popular mode). For my
4 GB RAM (Win7, 32bit) i can decode 54 characters in 8K19A, if i
remember correctly. In 16K25A i cannot even decode a single character.
When trying a lower number of characters with that given file you will
of course get no useful result but it is fine for searching the limit...
73, Stefan
Am 16.10.2016 20:01, schrieb [email protected]:
;-)now the light is on!
thanks Markus, I have to wait that my RAM grows a little in the mean
time will go haed with standard wspr, now testing with the lowpower
exciter driving the PA of the transverter. If all ok I should air a
stronger signal.
CU 73 Marco IK1HSS
----Messaggio originale----
Da: [email protected]
Data: 16-ott-2016 18.22
A:<[email protected]>
Ogg: Re: R: Re: LF: Idiot's guide to receiving Ebnaut? - Tutorial
part1
Yes the EbNaut decoder is indeed hungry for memory. I tried it once
(and never again) on a little 1 GB Atom netbook - all that happened was
that the harddrive light went on and the computer became unoperable
until hard reset. I normally use a heavier 8 GB laptop for this
purpose, but attempting to decode a stronger code (16K25) immediately
froze that machine as well.
Stefan's file with 8K19 ran ok also on a 4 GB tablet, although the
little1.83 GHz Celeron N2940 CPU chewed for 35 minutes to go through
all phase combinations. The default 30°-step setting finishes when it
says "25% complete", the finer 15° steps take four times as long.
Fortunately when the frequency is correct and propagation phase was
stable, the best decode usually appears already in one of the early
constant-phase trials.
73, Markus (DF6NM)
-----Ursprüngliche Mitteilung-----
Von: marcocadeddu<[email protected]>
An: rsgb_lf_group<[email protected]>
Verschickt: So, 16 Okt 2016 3:14 pm
Betreff: R: Re: LF: Idiot's guide to receiving Ebnaut? - Tutorial part1
Markus, Stefan,
thanks a lot for the nice tutorial, but I didn't succeed :-( either is
not enough idiotproof or there is something else..
What are the least system requirements for processing
Ebnaut decode?
In the shack I have a tower pc running XP, the clock is 1,99 Ghz with
456Mb RAM; I downloaded Ebnaut 5.0 and the audio file you provided and
after following the guide (some time later..) I got the message "not
enough memory to run the decoding" and indeed although the computing
resources were not under pressure (4%) I seen that a 1Gb swap file was
running so the answer should be add RAM.
I have also a netbook running w10 clock 1,7GHz with 1GB RAM: I tried
here but...
I'm unable to set thw Ebnaut window size and therefore the RUN button
is not visible, the desktop size /resolution cannot be changed so in
order to see the full Ebnaut window I have to switch in the vertical
mode (of course in this case I see all but the mouse command are still
the same so when I move in vertical the mouse the cursor move
horizzontal...). Anyway finally everithing was set and pushed the RUN
button.
The initialsation message was displayed but after 15 minutes all was
blocked with no sign of computing..
Did I miss something else?
Thanks for any feedback
73 Marco IK1HSS
----Messaggio originale----
Da: [email protected]
Data: 15-ott-2016 18.25
A:<[email protected]>
Ogg: Re: LF: Idiot's guide to receiving Ebnaut? - Tutorial part1
Hi Tony,
downloading from the Dropbox link appeared to work fine here. Anyway
I've temporarily ftp-ed Stefan's file to
http://df6nm.bplaced.net/VLF/fec_tests/10111913.wav
, hope that might help out.
One minor note: Riccardo's transmission
was announced and sent 0.6 milliHz above
nominal. Entering a frequency
offset of -14.9994 instead of -15 Hz produced a stronger and quicker
decode.
All the best,
Markus
-----Ursprüngliche Mitteilung-----
Von: Tony Baldwin<[email protected]>
An: rsgb_lf_group<[email protected]>
Verschickt: Sa, 15 Okt 2016 5:58 pm
Betreff: Re: LF: Idiot's guide to receiving Ebnaut? - Tutorial part1
Stefan
I can't seem to download the dropbox files, it just opens to a video
playback.
Tony
On 15/10/2016 11:58, DK7FC wrote:
Hi all,
There must be some progress.
I thought it could be useful to lern how to participate in EbNaut
detections by doing 2 exercises.
1: Generating wav files that contain the message
2: Configure the EbNaut decoder to read the wav file successfully.
Step 2 can be done today.
First you need to download the decoder software at http://abelian.org/ebnaut/
Then you need a wav-file containing a message. For this tutorial,
you
can have a wav file that was generated on my PC. It contains even 2
messages (rare!).
Download the file from:
https://dl.dropboxusercontent.com/u/19882028/LF/10111913.wav
To decode a EbNaut message you have to know the start time, the
symbol
length, the number of characters and the Coding. These informations
are
given by the TX stn.
Info for message 1:
Am 11.10.2016 19:04, schrieb IZ7SLZ:
Hello LF,
Riccardo IW4DXW informs that he also will be QRV EbNaut tonight on
137485.0006 Hz
Message contains 31 characters transmitted at 0.5 s/sym, coding
8K19A,
duration 14'40'' each hour starting 19.00Z till 23.00Z .
GL.
73, Domenico iz7slz
My file contains the 19:00 UTC transmission.
Info for message 2:
Hello Group,
considering the Stefan's request, i'm pleased to test again EbNaut
today on LF with a 51 characters message, 8K19A code, 0.5 s/symbol,
duration 22' 40''.
QRG = 137495.000 Hz
Scheduled UT: at hour, starting from 19.00 'till 01.00 of 12-10-2016
(included). Carrier will be "on" 5 minutes before the transmissions.
Now use the EbNaut decoder and select the wav file. For the first
message you can choose 31 characters, and 0.5 for the symbol period.
Then press start, just for a few seconds, then stop again. You will
now
see some more informations: The RX freqency and the file start time.
Now you can select the frequency offset, which is -15 Hz. Type that
into the field.
The next thing is the timing offset. The start time is 18:54:22.340
and the message started 19:00:00. You need to know the timing offset
precisely! If you don't get a decode, it is worth to vary the time
offset a bit. A timing offset could of course also come from the TX
stn, sometimes unknown. SpecLab causes a time offset of 4 samples. You
can also see the samplerate 117.19/sec. So the timing offset is: (4S)/
(117.19S/sec)+0.66sec+37sec+300sec = 337.69 seconds. Type that into
the
start offset field.
Now you can start the decoder and should get a decode quite quickly,
depending on your PC power.
By changing just the number of characters and the frequency offset,
you should receive the other message.
Fine, that was part 1 :-)
I need to get a feedback now if YOU got a successful decode or if
there are problems. Depending on that feedback i would then continue
with part 2.
Can someone of the 'newcomers' decode both messages?
73, Stefan
|
