From: WarmSpgs@aol.com
Message-ID: <c.5f80486.26696458@aol.com>
Date: Fri, 2 Jun 2000 15:26:16 EDT
Subject: Re: LF: Slow CW vs. BPSK etc.
To: rsgb_lf_group@blacksheep.org
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Rik posted while I was still struggling to finish typing mine.  If he had 
been a bit faster or I had been a bit slower, I could have addressed some 
part of the bandwidth issue. <g>

The necessary bandwidth of these methods may be correct as stated, but the 
actual radiated bandwidths may not be.  North American-style BPSK, a least 
that which is generated in LowFER transmitters, is not particularly 
spectrum-efficient.  It's a constant-envelope type, which means it can be 
amplified in high-efficiency stages (provided the output tuning isn't too 
sharp).  The extra sidebands are not a problem since we have 30kHz in which 
to play, and negligible radiated power to start with.  PSK31 gains its 
spectrum efficiency through envelope shaping, but that same requirement 
probably makes it a lot more susceptible to impulse noise than its 
brute-force BPSK counterpart.
 
As I mentioned in my previous message, I agree with Rik's assessment that 
narrower bandwidths with QRSS give us another window of opportunity for 
communication below signal levels where pure digital modes fall over the 
cliff edge.  The window is not limitless, because propagation-related phase 
shifts will eventually impact very narrow communication channels in the 
analog domain too.  Still, I think that window needs to be explored fully, 
and not dismissed out of hand just because it's slower or lacks digital 
purity.

73,
John