On Thu, 15 Apr 2004 10:01:50 -0500, Bill Hohnstein K HA wrote:
>Again, you're taking incorrect analogies.
No, I'm not. We're trying to separate a signal from noise. Our signal
may be narrowband (CW), but the noise is not. We can, of course, help
ourselves by reducing the bandwidth with narrow filters, but filters
ring, and if there's a lot of noise in the skirts the ringing is nasty.
If we make the bandwidth very narrow, the ringing is very close in
frequency to the signal masks it. That's why reception methods that
don't depend on narrow bandwidth can sometimes work better in the
presence of impulse noise. There are times, for example, where I get
better copy on a weak signal using DSP noise reduction than with a
super narrow filter. And, of course, it depends on the SHAPE of the
skirts.
So if we can figure out how to null the noise without nulling the
signal, that's a good thing. That's one of the ways that diversity can
work for us, but to use it, we have to know when the noise is precisely
in phase (so that we can reverse the polarity and cancel it) without
cancelling the signal.
Or, perhaps we can make the signal be coherent, making it add by 6
dB/doubling, but make the noise be un-correlated, so that it adds by
less than 6 dB/doubling. Note that Tom just said in a post that the
signal jumped out better if the two receivers were dead nuts on the
same frequency. That's coherent addition!
Some of what K9DX and W8JI have described is taking advantage of stuff
like this, whether they consciously know it or not. But by consciously
knowing it, we have the chance to figure out how to do it better!
73,
Jim K9YC
_______________________________________________
Topband mailing list
Topband@contesting.com
http://lists.contesting.com/mailman/listinfo/topband
|