On Fri, 29 Jan 1999 19:23:17 -0600, Jon Ogden <jono@enteract.com>
wrote:
<snip>
>>If you send dit-dit-dit-dah and noise or QRM causes you to
>>miss a dit and copy it as U instead of V, you have no way of knowing
>>you got it wrong.
>
>Yes, I do. Let's say the V was the start of the word Very. When I hear
>Uery, I know that I got it wrong. I have error correction.
Ok, if you're happy with "uery", no problem. I'm not.
<more snips>
>...but I have a hard time
>believing that anything can be pulled out BELOW a noise floor. That's
>why it is called the noise floor. Maybe the DSP helps to move that
>lower, but there is still a noise floor where the signal becomes
>unintelligible from the back ground white nosie. The DSP can give you
>some more dynamic range over the human ear. It can't work magic.
>
______________________________________________
Yes, it CAN work magic. It's done with signal averaging. The DSP
will record a short piece of signal, say a few hundred milliseconds,
and analyze it looking for a pattern buried in the noise. The pattern
of course, is the tone from the signal you're trying to receive. In
effect it "subtracts" the totally random stuff from the non-random
stuff and what's left is the signal. If you ever get a chance, watch
a modern digital storage oscilloscope in action. Put in a sine wave
which is totally buried in noise. In the "normal" mode, nothing is
visible to the eye except what appears to be white noise. Turn on the
"averaging" mode and like magic, the noise fades away and the sine
wave appears. The first time you see this happen it will give you
goose bumps. At least it did me.
This same technology can be applied to radio. NASA has used it for
years to dig spacecraft signals out of the noise. It works; now we
need it applied to amateur radio. It's coming.
73, Bill W7TI
--
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
|