Envelope-to: dave@picks.force9.co.uk
Delivery-date: Mon, 12 Sep 2005 14:57:29 +0100
Message-ID: <000401c5b7a1$56b3f8a0$0a8cf8d4@standalone>
From: "Andy" <actalbot@southsurf.com>
To: rsgb_lf_group@blacksheep.org
Cc: "Me Work" <actalbot@dstl.gov.uk>
Date: Mon, 12 Sep 2005 14:52:23 +0100
MIME-Version: 1.0
Subject: LF: Re: Re: Request for information.
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

John -
Yes please, if you could send the screen dumps and info to my  work email
address (its faster) -- actalbot@dstl.gov.uk

I'm surprised that you say the improvement with Africam was not that
significant.   Surely, the biggest problem with any coherent scheme like PSK
is maintaining symbol timing and carrier lock.  Certainly the timing can be
determined exactly, on a continuous basis, and most of the coherency issue
is resolvable too, so surely now that all the non-linear processing elements
of PSK reception have gone away shouldn't the net gain be potentially
massive ?

As I may have said at the time,  G0TJZ's chirpsounder monitoring software
makes use of the 1PPS from a GPS Rx fed to the soundcard and is  very
definitely a fully coherent system - albeit only to about 125us timing
resolution as a function of sampling at 8kHz

Andy  G4JNT


-----Original Message-----
From: John Andrews <w1tag@w1tag.com>
To: rsgb_lf_group@blacksheep.org <rsgb_lf_group@blacksheep.org>
Date: 2005/09/12 13:58
Subject: LF: Re: Request for information.


>Andy,
>
>Bill de Carle produced a version of Africam that allowed GPS sync. As I
>recall, we thoroughly tested it in early 2004 with signalling rates as slow
>as MS2000 (0.5 b/s). At the lower data rates, it provided a signal lock
much
>more quickly than the non-GPS version. It also held a signal better through
>deep fades.
>
>That said, the performance gain was not dramatic. WOLF was more effective,
>though in a wider BW. As I recall, there were five of us capable of
>transmitting and receiving GPS-sync Africam, and a couple of other guys
>would copy the signals using the traditional sync. Jay Rusgrove and I have
>continued to run it from time to time. But as Africam is a DOS based
>program, and can be fussy about sound cards (most of us were using Bill's
>S-D board), there hasn't been much interest.
>
>The GPS-sync concept is still appealing, however. As last discussed, the
>next idea was to feed 1pps ticks into the right channel of a sound card. I
>don't believe that this has been implemented, however. It would be
>interesting to see how that worked with an MS1000 version of WOLF. The
>quicker sync time might compensate for the lower data rate.
>
>I should still have email messages and screen dumps left from the Africam
>tests, if you have any interest.
>
>John Andrews, W1TAG
>
>
>