inmarsat

Can anyone help me? I can't find enough information about standard
Inmarsat-C. I found information about Inmarsat-C in the group
FAQ-Inmarsat, but this information isn't correctly. I built Matlab
receive model and I use Matlab BPSK demodulator. I tryied with this
demodulator, afterthat I changed it with Matlab DBPSK demodulator, but
results were mostly same. I didn't see synchroword and structure of
Inmarsat-C isn't correctly. Are synchroword and frame structure
correct? Can anyone to send me more information about my problem?
I'm familiar with standard Inmarsat-C, but I want to know more
information about structures of Inmarsat-C frame.


PS: My English isn't good. My e-mail:crusaders792002@yahoo.com

Hello,

My experience is with Inmarsat B, M, mini-M, and Aero. Basically if you
don't have the specs, you are pretty much sucking wind. In the
protocols that I worked with, the frame structure was channel
dependent. About the only constant is the signal unit structure, even
though there is an extended version in Aero. You will want to get the
spec from InMarsat. Also Inmarsat uses both assymetric and symmetric
forms of DBPSK modulation depending on which frame type you are looking
at.

Clay

Crusader Wrote via direct email

>>Hello. Thank you very much for You answer, but I have got some
>>questions. Sometimes I use a software of company WaveCOM and Hoka(Code
>>32). In the software of Hoka(Code 32, last version) I can use two
>>universal demodulator: BPSK and 2DPSK. But frame structure of input
>>signal, I can see only when I use 2DPSK demodulator.In the Wavecom
>>software, I don't see demodulator output. I see only final results. Do
>>you use these software applications?

No I don't use these. I wrote my own code in Moto56300 assembly. Some data 
uses QPSK modulation, and I doubt that your canned 2DPSK routine can handle 
the symmetric DBPSK modulation.

>>Do you know where I can find maybe free source code of Inmarsat C or
>>some amateur applications about this standard? Do you know, after
>>demodulator there is Viterbi decoder(that's structure there are in
>>WaveCom software) or structures: DeInterleaver->Viterbi->DeScrambler?

Almost always a Viterbi decoder is used. However some of the newer data 
channels use turbo coding instead (mini-M). And after the error correction 
there is both a bit scrambler and a CRC check. I don't know about free code 
for the whole thing. You should be able to find some of the Viterbi stuff 
though. Much of Inmarsat uses rate 1/2 K=7 TCM.

>>I have a small information about Inmarsat-B,but that standard is next
>>standard after Inmarsat C. I know Inmarsat-B include voice data.That's
>>right?

I have never worked with Inmarsat C, and yes B and M and mini M all support 
voice.

IHTH,
Clay

Hello. Thank you very much for Your answer.

Hello!
Can I ask you a new question?
I don't understand  next fragment of Your first message:" Also Inmarsat
uses both assymetric and symmetric
forms of DBPSK modulation ". I know there are:BPSK and DBPSK
demodulators. In BPSK we don't use prevous phase of the PSK
signal(prevous phase value). In DBPSK, we use prevous phase value.
Thank you for advice!

Hello Crusader,

In assymmetric DBPSK, a one bit means to change phase by 180 degrees
and a zero bit means no change in phase. The assymmetry stems from
there not always being a change. I.e the error functions for ones and
zeros are not the same.   In symmetric DBPSK, a one bit means to change
phase by +90 degrees and a zero bit means change phase by -90 degrees.
Thus there is always a change of phase from symbol to the next symbol.
Hence the error functions are the same.  Even though the stream overall
contains 4 phases, this is called BPSK, since for any symbol period
only one of two phases is allowed. So you may think of this as a
special case of QPSK. By the way Inmarsat also uses regular QPSK. Are
you trying to decome the bulletin board channels?

Clay

Hello, Clay!
According to me, the problem is in my demodulator, but I'm not sure. I
have an
experience in Inmarsat-A. In a lot of documents about Inmarsat-C, I
read :"Demodulator BPSK". Sometimes I use commersial PC DSP board(I
don't know what is called, that DSP board is a very old, i.e. a
software of DSP board), but there are very good carrier PLL and Timing
PLL. It demodulator is DBPSK(after a few expirience I fixed that,but in
the description of software of PC DSP board, the builder of that
software called 2-PSK demodulator ). In my Matlab model, there is DBPSK
demodulator. When I use that demodulator, there are some arrange(i.e by
colums and rows, I'm not sure about that arrange). There are  162 bits
of row(according to standard
Inmarsat-C).But when I use BPSK demodulator, there aren't any arrange.
I want to tell you, Matlab DBPSK demodulator is same with PC DSP board
demodulator(i.e demodulator's output). I wrote application for
Inmarsat-A where I used DBPSK
demodulator,but Inmarsat-A use BPSK demodulator. I tryied with Matlab
BPSK
demodulator and results was almost always same, except few errors. I
have not any problems with that standard. I don't know where I mistake.
I give you next link :www.soumu.go.jp/joho_tsusin/eng/
Resources/Legislation/MRA/pdf/1990_no.566.pdf
 about Inmarsat standards. I have got that document. Can you tell me
does document is correct, i.e. I know about you experience Inmarsat-B
and etc.? Can I believe it? Do have any problems with Inmarsat-B and
etc.?

I forgot. If can't find that PDF file, you can write in Google the name
of PDF file.

Hello Crusader,

I looked at the document that you have, and it looks consistant with
the spec, however what you have is very brief. In the real spec, each
of the sections gets explained in detail. Plus your document, omits the
structure of the signal units.

Do you have a real world signal you are trying to decode? When I did my
work on this, I had access to functioning equipment to verify my work.

Clay