Topics

Test version 5 of the Multipsk version 4.43.1 - Improvement of the EGC decoding


tseug
 

David, thanks, I meant to post the Inmarsat C NCS freqs within the thread for convenience:

1537.1000MHz, NCS, ALPHASAT (4AF4), LES 344, Region IOR (EMEA)
1541.4500MHz, NCS, Inmarsat 3F5, LES 144, Region AOR-E
1537.7000MHz, NCS, Inmarsat 4F3, LES 044, Region AOR-W (AMER)
1541.4500MHz, NCS, Inmarsat 4F1, LES 244, Region POR

with the active satellite info but obviously messed up whilst cutting & pasting a 'long' post on my smartphone :(

Another indoor antenna option would be to take one from an old 'Worldspace' radio. Maybe find something like that at a a car boot sale / garage sale though they weren't very common.


David L. Wilson
 

You probably want to mention the post with the 1545 MHz frequencies are the Aero frequencies rather than INMARSAT-C (as it appeared under the Improvement of EGC decoding thread).  I am not sure if I mentioned it hear but my list of identified on AMER/AOR-W/AOR-E satellites is at www.reddit.com in r/signalidentification.  It includes INMARSAT-C, INMARSAT Aero, and INMARSAT IsatM2M/D+ (and I think is the only source of the IsatM2M/D+ from my personal decoding via various steps and know of no otherwise available software to do that). 

Also, sometimes INMARSAT panel type antennas appear on eBay.  There are 2 there now with one prices more than twice the other.  I have one of these myself and with a pre-amp it works OK (better than all those mentioned as I have all of those also) but I have a 1 meter dish that is of course better than it.


tseug
 

Hopefully this won't class as 'off topic'

When reading Patrick's guide note that the satellite info is currently as follows:

SATELLITE ID = 0 (Long 99W) Seq = 6 Psmc1 = 1545.0200MHz Psmc2 = 1545.2050MHz
SATELLITE ID = 1 (Long 54W) Seq = 5 Psmc1 = 1545.0150MHz Psmc2 = 1545.1950MHz
SATELLITE ID = 2 (Long 142.5E) Seq = 4 Psmc1 = 1545.2000MHz Psmc2 = 1545.0100MHz
SATELLITE ID = 3 (Long 24E) Seq = 3 Psmc1 = 1545.1900MHz Psmc2 = 1545.0050MHz
SATELLITE ID = 5 (Long 144E) Seq = 2  Psmc1 = 1545.1050MHz
SATELLITE ID = 6 (Long 25.5E) Seq = 1  Psmc1 = 1545.1150MHz
SATELLITE ID = 7 (Long 97.5W) Seq = 0  Psmc1 = 1545.1100MHz
 
However, as previously mentioned the system runs as a Virtual Inmarsat I3 Network (to maintain compatability with old mobile equipment) overlaid onto the Inmarsat I4 System using 
the following actual satellites
 
SATELLITE ID = 0, Inmarsat 4F3 (*) Seq = 6 Psmc1 = 1545.0200MHz Psmc2 = 1545.2050MHz
SATELLITE ID = 1, Inmarsat 3F5 (Long 54W) Seq = 5 Psmc1 = 1545.0150MHz Psmc2 = 1545.1950MHz
SATELLITE ID = 2, Inmarsat 4F1 (*) Seq = 4 Psmc1 = 1545.2000MHz Psmc2 = 1545.0100MHz
SATELLITE ID = 3, Alphasat (Inmarsat 4AF4) (*) Seq = 3 Psmc1 = 1545.1900MHz Psmc2 = 1545.0050MHz
SATELLITE ID = 5, Inmarsat 4F1 (Long 144E) Seq = 2  Psmc1 = 1545.1050MHz
SATELLITE ID = 6, Alphasat (Inmarsat 4AF4), (Long 25.5E) Seq = 1  Psmc1 = 1545.1150MHz
SATELLITE ID = 7, Inmarsat 4F3 (Long 97.5W) Seq = 0  Psmc1 = 1545.1100MHz

You asked about patch antennas - 
rtlsdrblog also market a patch antenna (worst performance out of all those I mention here BUT it's already waterproof and contains an LNA plus if you want to look at Iridium it will do for that as well).
nooelec make a microstrip patch which will work
sdrkits market a modified 'gps' puck which is suitable.
Better than all of those is the home build design for a simple airgap patch by Adam, 9A4QV, and if you want to get a touch more gain then a home made airgap based on the outernet patch will give you that.
The frequency response of each of the above is not exactly the same
Another option would be a small home-made helical (see uhf-satcom L-band lots of useful info about satcom on the other pages there as well but bear in mind the frequency listings are old - another very good source of satcom info is the long-established irc channel 'hearsat' which is currently on the freenode network)

For an LNA I'd look at the nooelec sawbird IO or sawbird IO+ to give filtering and amplification rather than a wideband LNA


WB9EZB
 

Thanks 


On Jun 27, 2020, at 12:46 PM, Patrick Lindecker <f6cte@...> wrote:



Hello,

 

For EGC, a patch antenna and preamp is sufficient.

 

Look at this link: http://f6cte.free.fr/EGC_easy_with_Multipsk.pdf

 

73

Patrick

 

De : multipsk@groups.io [mailto:multipsk@groups.io] De la part de WB9EZB
Envoyé : samedi 27 juin 2020 18:56
À : multipsk@groups.io
Objet : Re: [multipsk] Test version 5 of the Multipsk version 4.43.1 - Improvement of the EGC decoding

 

I am reading about everyone listening to inmarsat. I am located 89 west 45 north.
could I receive it using a patch antenna and preamp. I can’t put up a dish.
any recommendations for antenna and preamp. Running Multipsk and SDR play
radio. Thanks 




Avast logo

L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast.
www.avast.com




Patrick Lindecker
 

Hello,

 

For EGC, a patch antenna and preamp is sufficient.

 

Look at this link: http://f6cte.free.fr/EGC_easy_with_Multipsk.pdf

 

73

Patrick

 

De : multipsk@groups.io [mailto:multipsk@groups.io] De la part de WB9EZB
Envoyé : samedi 27 juin 2020 18:56
À : multipsk@groups.io
Objet : Re: [multipsk] Test version 5 of the Multipsk version 4.43.1 - Improvement of the EGC decoding

 

I am reading about everyone listening to inmarsat. I am located 89 west 45 north.
could I receive it using a patch antenna and preamp. I can’t put up a dish.
any recommendations for antenna and preamp. Running Multipsk and SDR play
radio. Thanks 




Avast logo

L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast.
www.avast.com



Patrick Lindecker
 

Thanks to all the testers for the positive reports of the last test version.

 

73

Patrick

 

De : multipsk@groups.io [mailto:multipsk@groups.io] De la part de David L. Wilson via groups.io
Envoyé : samedi 27 juin 2020 16:43
À : multipsk@groups.io
Objet : Re: [multipsk] Test version 5 of the Multipsk version 4.43.1 - Improvement of the EGC decoding

 

I compared it with the pay one from www.inmarsatdecoder.com.
In my case, with that inmarsatdecoder, I use the waterfall display from the program Argo to help tune it and ist is still not easy.  But do not need that with MultiPSK as MultiPSK tunes much much easier by clicking on its display. 
I do not have Tekmanoid but some LES info on its webpage is wrong (like 103--it is Yamaguchi, Japan not Russia---identified in both INMARSAT literature and from monitoring verified). 
I never experimented much with ScytaleC as it did not have a reasonable interface for output data when I last looked at it.




Avast logo

L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast.
www.avast.com



WB9EZB
 

I am reading about everyone listening to inmarsat. I am located 89 west 45 north.
could I receive it using a patch antenna and preamp. I can’t put up a dish.
any recommendations for antenna and preamp. Running Multipsk and SDR play
radio. Thanks 


David L. Wilson
 

Yes, indeed, I think I mentioned somewhere about lookup tables being used and it is amazing how other software has errors due to being wrong information from start in theirs (not just not being up-to-date),  INMARSAT has used the term "virtual satellite" in regard to 99W and 97.5W and treats them as two different satellites in bulletin board identifications and even lists those slightly different locations.  Additionally, as you probably know (this post is really for others), INMARSAT operations (C, Aero, ISATM2M) are kept separate on them for AMER and AOR-W operations.  I suspect that is being done so a possible future move to separate satellites will be user transparent.  Additionally, what they call "harmonization" has combined some operations that were at separate LES but the same satellite in the same service into one operation from one uplink station but the old LES numbers are still used on their old frequencies--again to make things easier for the user.  For example, much is uplinked from Laurentides, Canada, but you likely never will see a LES number that identifies that.   I wish those of us  with this detailed interest could find a way to communicate outside of here as I agree it is a little off topic.  I did put my recent list up on reddit at r/signalidentification/


tseug
 

For multipsk purists, stop here and move to the next thread as this is becoming a bit marginal.

LES info in all of the software is done via lookup tables and both scytale-c and tekmanoid need updating (though it's not really as simple as that as even Inmarsat can't decide whhich is which from one document to the next and the 'names' of the land earth stations also vary). Both software authors are aware yet neither have pushed out any updates (not as responsive to user input as Patrick....). In terms of decoded output the one from the website you mention is by far the best...but it's also the most expensive.
People still talk about the 99W and 97.5W satellites but that makes no real sense as the channels for those are from a single sat (4F3)...in the same way 24E and 25.5E are off a single sat (Alphasat). Even the LES info is somewhat redundant as the traffic can (and does) go via fibre-optic links to whichever uplink station best suits it at the time.


David L. Wilson
 

Also note that "technically" what we are calling LES are LESO (LES Operator) operating in global coverage.  This is done by  uplinking earth states remotely to be able to reach earth stations worldwide.


David L. Wilson
 

I compared it with the pay one from www.inmarsatdecoder.com.
In my case, with that inmarsatdecoder, I use the waterfall display from the program Argo to help tune it and ist is still not easy.  But do not need that with MultiPSK as MultiPSK tunes much much easier by clicking on its display. 
I do not have Tekmanoid but some LES info on its webpage is wrong (like 103--it is Yamaguchi, Japan not Russia---identified in both INMARSAT literature and from monitoring verified). 
I never experimented much with ScytaleC as it did not have a reasonable interface for output data when I last looked at it.


tseug
 

Tekmanoid gives the BB data but as a table rather than in the tabbed display area.
As for the tuning, all three (and stddecoder) work best at the same tuned point but all have differing autotuning responses. Scytale-c seems to need a ridiuculously wide bandwidth to actually tune well. The dedicated decoders all give much nicer output and far more info...but that is to be expected. Multipsk is doing a nice job now and would make an acceptable decoder for people who only want the very limited EGC messages....and as we all know, Patrick is constantly adding value to his programme ;)


agene
 

Hi Patrick,

Your new test version 5 seems to be working very well.

I tested today with my RSPdx using 3 VRX, all on 1541.45 MHz, each VRX connected to a different EGC software program,
1. MultiPSK
2. ScytaleC
3. Tekmanoid

In my opinion, MultiPSK gave the best result, ScytaleC needed to be tuned to 1541.445 MHz and then drifted a bit more. Tekmanoid worked fine, but gave no LES data.

Of interest, you now have MultiPSK giving message ID in Octal, ScytaleC gives it in both Decimal and Hex and Tekmanoid does not give the raw data at all.

Regards, Mike Simpson
South Penrith, NSW, Australia



tseug
 

Tested again on the Alphasat EMEA NCS and the full Buletin Board list of channels is correctly displaying the LES numbers


Patrick Lindecker
 

Hello to all,

 

I propose this test version 5, which fixes a problem about the EGC LES number: http://f6cte.free.fr/MULTIPSK_4_43_1_test_version_5.zip

 

Paste this address in your Internet Explorer or equivalent. Download the file.
Extract the ZIP file and replace the “official” Multipsk.exe with this test version. Before, make a copy of the “official” Multipsk.exe, in case of.

 

Thanks to your reports to the group.

 

73

Patrick

 




Avast logo

L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast.
www.avast.com