From: "james moritz" <j.r.moritz@herts.ac.uk>
To: rsgb_lf_group@blacksheep.org
Date: Sun, 5 Dec 2004 14:45:49 -0000
Message-ID: <000001c4dad9$1e7acfe0$e6a4c593@RD40002>
MIME-Version: 1.0
Importance: Normal
In-Reply-To: <41B0ECBF.1090900@freenet.de>
Subject: RE: LF: Variable-PSK theory (warning - goes on a bit!)
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

Dear Wolf, LF Group,

-----Original Message-----


...Jim wrote:
>> I beleive Wolf has used a type of PSK where the phase of the carrier 
>is varied linearly from 0 to pi radians,
>> or pi to 0 radians, for a 0 to 1 or 1 to 0 bit transition.
>Yes that's true, but the transition did not use 100% of the symbol time 
>(but 10..20%) to stay compatible with the "PSK31" standard decoders. So 
>this was not (yet) comparable to MSK.

OK, I did not realise that - I guess the effect would be to reduce the
amount of power in the quadrature component significantly, but also to
increase the amplitudes of the high-order sidebands significantly. But I am
pretty sure the tests I did a couple of years ago did use a phase transition
period of 100% of the bit period - this seemed to work quite happily with
the "coherent" BPSK decoder. I don't see any obvious reason why the extra
quadrature component in the variable-PSK signal should prevent the standard
PSK31 demodulator from working with "transition period = bit period", but
then I don't know a lot about how the demodulator actually does work, so I
could well be wrong! 

>I have modified G3PLX's QPSK-mode into Offset-QPSK by delaying the 
>Q-component in the modulator by a half symbol time, and in the 
>I-component of the demodulator to compensate this. By doing so, there 
>are no more 180°-phase transitions in the signal any longer (only +/- 
>90°), so the amplitude only drops by 3dB. Still not perfectly suited for 
>class C amplifiers, but the requirements for a "linear" TX are not as 
>tough as for AM-shaped QPSK or BPSK. No tests done with any of the 
>"Novel" (G)MSK decoders yet, but at least I am beginning to understand 
>what the tricky parts are.

Will be interested to see what you come up with. BTW, thanks for the CW QSO
earlier today!

Cheers, Jim Moritz
73 de M0BMU