Hello, thanks a lot for information, ok some answers:
First why am I wanting navtex in fldigi ? - I am under linux, sorry that so evident for me ;-) may be I did not mention it, and appart from the java decoder, do not find anything else for navtex - second, it is GPL, so if needed it s easy to have a look how something is done, and eventually modify it ( that s the hard part) - third fldigi, is very flexible, and I have only one ssb and soundcard, and single handed, so if I want to have automatic reception plus freq change that would be a good idea, and with xml rpc it seems feasible I don't yet have pass the exam to have a Ham licence (shame on me, but I think I am not the only sailor in this situation, so for the moment no pskmail) second, for the decoding you are right using rtty, may not be the best idea, but it should allow me to have a sequences of 1 and 0, and from there I will know what to do, I don't know a lot about signal processing, but I know a little programming I can use a fprintf at many place in rtty.cxx, for the moment I am trying there : > case RTTY_RX_STATE_DATA: > fprintf(stderr,"%d",bit); ok for the moment, I do not yet succeed to have access to the "good signal, in this long list of 0 and 1, I have assumed that this list of 1 and 0, repeated as long as the signal last, will after chunking findind how long is a bit, will lead to the right amount of 1 and 0, but no, not yet) And it's a fun introduction (it's like solving a riddle) for me to understand better fldigi... and digital mode and may be in the future being able to help in development. thanks for your atention, and clues, and if anyone have an idea thanks in advance, I may now have a look to signal processing in fldigi to understand better, where this list of 1 and 0 come from... and why it is not the right one... If someone else is interrested, I can give all the data I already have, (sound sample with various level of noise, transcriptions, and other data ...) Best to all, Gabriel On Fri, Apr 23, 2010 at 1:14 PM, M. Jenssen <[email protected]> wrote: > *Hi Gabriel,* > *to complete your statement, here a short summary about AMTOR/SITOR and > FEC as compiled by Patrick (author of Multipsk):* > *quote* > ** > ** > ** > Description : > Baud rate : 100. A character is composed of 4 « space » and 3 « mark » > bits. A block of characters (called « Block 1 » or « Block 2 ») is composed > of three characters (total duration: 3 x 70 = 210 ms) and a pause of 240 ms > duration. During this pause, the receiving station emits a control signal of > 70 ms duration to say « please next block » (symbol "ACK" for acknowledge) > or to ask for a repetition of the last block, in case of erroneous character > (symbol "NACK" for not acknowledge) > Speed : between 0 and 67 wpm (67 wpm corresponding to the 50 bauds RTTY) > depending of the number of repetitions > Modulation : FSK two tones (« mark » and « space », "mark" high) with a > shift between tones of 170 Hz > Reception mode : USB > Character set : CCIR476-4 (32 letters, 32 figures and 6 control, idle or > phasing signals) with no error reset character (« fixed length » > characters). > In SITOR A, the character set may be extended to the Cyrillic characters by > using the character for « un-perforated tape » (third shift). > In AMTOR A, the character set may be extended to the ASCII characters (32 > to 128) by using the character for the « un-perforated tape » > Shape of pulse : rectangular > Bandwidth : about 800 Hz (due to rectangular shape), > Demodulation : non coherent > Synchronization : automatic using the signal > Correction code : yes via automatic repetition and easy detection of an > erroneous character due to the particular content of a character (4 > « space » and 3 « mark » bits) > Convolution code : no > Interleaving : no > Pmean/Ppeak : 1 > Note : In AMTOR ARQ LISTEN, characters are tranlated the best possible > way because no repetition can be requested. Therefore, receive quality is > not generally very good. > > - AMTOR FEC (called also SITOR-B for the navy (as a maritime mobile > service system) or simply FEC) collective B-mode > > <http://f1ult.free.fr/DIGIMODES/MULTIPSK/amtor_fichiers/AMTOR_FEC.mp3>** > > Description : > Baud rate : 100. A character is composed of 4 « space » and 3 « mark » > bits. It is an uninterrupted stream of characters, contrary to AMTOR ARQ. > Speed : 71 wpm > Modulation : FSK two tones (« mark » and « space », "mark" high) with a > shift between tones of 170 Hz > Reception mode : USB > Character set : CCIR476-4 (32 letters, 32 figures and 6 control, idle or > phasing signals) with no error correction character (« fixed length » > characters). Small characters are transformed in capital characters and > special characters as « é » are transformed in standard capital (in the > example: « E ») > Shape of pulse : rectangular > Bandwidth : about 800 Hz (due to rectangular shape), > Demodulation : non coherent > Synchronization : automatic using the signal > Correction code : no > Convolution code : no > Interleaving : each character is repeated 5 positions later (that is 350 > ms between 2 transmissions of the same character). This diversity in time > allows the correction of an erroneous character. Its detection is easy due > to the particular content of a character (4 « space » and 3 « mark »). > Pmean/Ppeak : 1 > Lowest S/N : about -5 dB > Note : There are 2 FEC modes: the collective B-mode (described before) and > the selective B-mode (little differences as an inversion of bits). > > *unquote* > > *You will not have much fun in using rtty mode for reading Amtor/Sitor.* > *In any case, you can read this mode with MixW and Multipsk. I personally > do not see any advantage in fldigi having this mode incorporated as it is > used sporadically only on our bands and as it does not offer any better > performance than those we got at our disposal. When I want to read Navtex - > when we got bad wx in our region - I just use one of those programs.* > *vy73 Fred DL6XAZ* > ** > > _______________________________________________ > fldigi-alpha mailing list > [email protected] > https://lists.berlios.de/mailman/listinfo/fldigi-alpha > >
_______________________________________________ fldigi-alpha mailing list [email protected] https://lists.berlios.de/mailman/listinfo/fldigi-alpha
