You also have to look at the Eb/No required to support
a given bit density without error. Digital modulation
techniques like QAM support many bits/hz (i.e. are very
spectrally efficient), but require a higher signal to noise
ratio to work properly. The power efficiency of the
modulation scheme will determine for a given link (fixed
carrier power, path loss, and noise density),
which scheme will more pass more bits in a given amount
of time. Perhaps voice has a higher spectral efficiency,
but also has a higher error rate for the same Eb/No, thus
rendering the effective bits/unit spectrum the same
at low Eb/No's. In other words which modulation technique
get you closer to Claude Shannon's limit?
Anecdotally, it would seem that when conditions are marginal,
one can pass more information/unit time using CW than voice.
Have you ever exchanged "rifle shots" "22" "bang bang"
with some equally PW guy on an SSB list net. Some of those
exchanges can go on forever!
Also, the argument Vic makes can go both ways. Q signals,
abbreviations, and contextual clues can be used for data
compression in both CW and Voice communications.
On/Off (CW) keying is certainly still viable modulation
scheme by virtue of its simplicity. It is still used in
all kinds of industry applications. If I am not mistaken
NASA is using it in some of their earth to space optical
(laser) communications systems. I suspect their coding scheme
is something other than international morse, but nevertheless
CW remains an attractive modulations scheme. BTW, the simplicity
of the equipment required is even greater when there is a trained
pool of operators carrying decoders around in between their
ears. This allows the equipment designer to throw away the
all that digital stuff - grin grin.
73 de Mike, W4EF............
----------
From: Vic Rosenthal[SMTP:rakefet@rakefet.com]
Reply To: rakefet@rakefet.com
Sent: Thursday, January 28, 1999 4:04 PM
To: Peter Chadwick
Cc: 'amps'
Subject: Re: [AMPS] Re: SERIOUS commentary from N4XY on "no-code"
Peter Chadwick wrote:
>
> >Given the fact that cw is far more spectrum efficient
>
> It isn't. Spectrum efficiency is bits/Hz.
>
> HF packet is about 0.1 bits/Hz. HF CW is about 10bits/sec, and needs about
> 50Hz to allow for fading etc, so is about 0.2 bits/Hz. Analogue speech is
> often reckoned as about 2 bits/Hz.
Well, it depends on what the meaning of 'bit' is (hmm, where have I heard that
before). In my opinion, the word 'yes' is one bit. Just because you are
transmitting a lot of bits in your sense doesn't mean you are communicating a
lot of intelligence.
73,
Vic, K2VCO
Fresno CA
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|