"Dick Carroll;" wrote:
Floyd Davidson wrote:

It would be if I'd said it! But YOU said it! I know better. I also
know -right alaong with every active ham- that you increase SNR by
*narrowing* bandwidth. Seems it gets rid of some of the noise while
preserving the signal, assuming of course the signal is narrow enough to
fit within the narrowed bandwidth. As it happens, a CW signal is!

We were discussing how to get increase efficiency, not how to
demonstrate a 0 bit error rate for the slowest possible bit rate.

If narrowing the bandwidth does reduce the SNR, then you were
not making effective use of the channel bandwidth to start with,
and must be using a relatively inefficient mode of operation for
the conditions you have chosen.

On the other hand, if you are making effective use of the
bandwidth, reducing it is merely going to make your
communications *less* efficient by reducing the channel capacity
and thus causing you to pass less information.

Knowing Shannon's theories would make all of this much easier for
you to understand. Here, I did put it up on my web page, so that
you can download it without have to embarass yourself asking for
it in email. I said it was 500K... but I'd misplaced a decimal
and it is more like 5.7Mb in size.

http://web.newsguy.com/floyd_davidson/techie/shannon.pdf

--
Floyd L. Davidson http://web.newsguy.com/floyd_davidson
Ukpeagvik (Barrow, Alaska)