In article , Floyd Davidson
writes:

Mike Coslo wrote:
Dick Carroll; wrote:

Floyd Davidson wrote:

some snippage

You increase the SNR, regardless of the bandwidth, by increasing the
signal level DICK.


Well, so much for your technical knowledge if THAT"S all you know about
it. Any experienced
ham, even without ANY tech schooling whatever, knows better than that.

As a dilletante, I realize that in any ratio, there are two numbers.

Actually, there are three (bandwidth, signal, and noise) which
are related to channel capacity by the following formula

Capacity = Bandwidth * Log2 ( 1 + Signal/Noise )

The debate is over comparing *efficiency* of different modes (CW
and PSK-31), and hence the channel capacity for such a
comparison, must be normalized.

Reducing the Bandwidth parameter does decrease the observed SNR
in the channel, but the Capacity is not increased because the
actual noise power per Hz is unchanged.

Ahem. Four. Include error rate. :-)

Claude Shannon used teleprinters as a working example in his 1947
landmark paper.

That made it more familiar to communications people in the REST of
the communications world. Not many radio amateurs knew how
teleprinters worked or how they were coded in 1947. :-)

"Shannon's Laws" apply to EVERY communications medium, wired or
wireless. According to a few ignorant extras the "don't apply" to amateur
radio. :-)

So while it is quite possible to make the s/n ratio larger by
increasing the signal, it is equally possible, and sometimes much better
to increase the s/n ratio by lowering the noise. Sometimes it is the
*only* option available.

However, what has to change is the noise power per Hz, and
reducing the bandwidth does not change that.

Heh, it's hard enough to get amateurs to use the proper multiplier
prefix on frequencies, let alone grasp a concept of noise power per
unit bandwidth. :-)

Seems like narrowing the bandwidth might just do that!

Increasing the signal power has the desired effect. There are
other ways to accomplish that, of course. Reduction of noise by
any means other than reducing the bandwidth (switching from an
omni directional antenna to a directional antenna, for example)
will have the desired effect.

your humble hockey puck, 8^)

Hows come, then, you don't have a Canadian call sign?

He might have drunk all his Canadian Club. :-)

LHA