Content-Transfer-Encoding: 8bit
Message-ID: <3617AC3245C2D1118A840000F805359C01AB8C85@pdw-mercury-1.dera.gov.uk>
From: "Talbot Andrew" <ACTALBOT@dera.gov.uk>
To: rsgb_lf_group@blacksheep.org
Subject: LF: Slow CW vs. BPSK. & Computer modes
Date: Fri, 2 Jun 2000 09:21:17 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>


>This is all-important. Extremely weak stations require considerable 
>additional work by the 'computer between the ears' to decipher 
>what is signal and what is noise on the screen - just like aural 
>Morse, but visual. This accounts for several extra dB of gain, and 
>however slow you send the Morse, this advantage is still available. 
>It would take a very sophisticated computer to be as good. 

NO !  With these modes the computer is not doing anything to aid
extraction of the signal that could not be done with conventional
hardware.    The 'several extra dB gain' is just there because more
noise has been filtered out.  The brain is just looking at amplitude
changes in this narrow bandwidth and integrating over the signalling
period (looking at brightness over the length of a dot). 

All Spect....  is,  is a bank of narrow filters.   There is nothing in
theory to stop you making a crystal filter, or a whole bank of them this
narrow.   For a display mechanism, a chart recorder - multiple pen ones
are available.    All this technology was available in the 1950s as I'm
sure those who were around in that era will remember.

All the computer does is make this much simpler and cheaper and
available to everyone, it is only a filter and display mechanism so
please can we stop referring to Spect...   as computer modes.   It can
be done in other ways.

Coherent etc and PSK31 are computer modes,  SPECT..... is not, it is
only a filter bank and display.

Real computer / DSP modes mean making use of coherent detection and
error correction - and have nothing to do with the speed of the
signalling.   

There seems to be no appreciation on this newsgroup of the real value of
coherent detection - that is, having phase information available when
the signal is decoded and coherently locking to the signal carrier and
bit  timing.   Simple theory, as I covered in a previous email, shows
mathematically the considerable theoretical advantage of coherent
detection over non coherent, there is no dispute about this.

SO unless your ears / eyes can respond to signal phase, is doesn't
matter how much personal pride says that the brain is a marvelous
computer, is will never compete against PSK  :  

Given the same TOTAL overall signalling speed and comparing like with
like on the end to end link.

Repeating CW characters time and time again has to be compared with true
error correction and low data rate signalling.

Computer error correction does the same job as the eye interpreting
between the dots on Spect...,   and with well chosen codes can do an
awful lot better.  These optimum codes are only just beginning to appear
thanks primarily to the huge research effort funded by the mobile phone
industry, but hopefully Coherent etc will soon make use of them - the
equations are public information.  The maths is similar, just the
implementation different.

Coherent etc and PSK31 are computer modes,  Spect... is not.

Andy  G4JNT


-- 
The Information contained in this E-Mail and any subsequent correspondence
is private and is intended solely for the intended recipient(s).
For those other than the recipient any disclosure, copying, distribution, 
or any action taken or omitted to be taken in reliance on such information is
prohibited and may be unlawful.