[Discuss-gnuradio] Re: information theory -- follow up

[MacLeod, Matthew]

> Eric -- I haven't been here very long, but my assumption about this code
to
> demod two FM stations at once is NOT an example of how you can demod two
FM
> stations "on a single frequency" as stated by David Weinberger in this
Salon
> article.

I had that basic problem with the article as well.

But my main problem is the colour analogy. It doesn't really account for a
lot of things. If you're looking at a 60W blue light bulb flashing signal at
you, then someone turns on a 1 000 000W blue light source behind it, you're
not going to be picking up much useful information any more, now are you?
Even if the two lights were different colours it would be difficult.
Basically any situation where you get washed out images, glare, partial
reflections, or any other visual problem show that just because colours
behave mostly like radio waves, and that we can still get usually get useful
information from them, doesn't mean there's infinite capacity there.

Also consider what some people do with university crib sheets. To get more
on the page they will sometimes write in two different colours, and wear 3D
glasses to let them separate the green and the red, or whatever colours they
choose, by closing one eye or the other. But without the glasses it's really
quite hard to read. All the lenses really are are band pass filters, which
we already do with radio waves.

Although I agree with a lot of the arguments, I don't think the colour
analogy is all that useful, and doesn't hold a lot of insight. I like the
multiple voices idea a lot better, which is more akin to CDMA, or how
freeing the spectrum gets rid of losses to guard bands.

What I would also see as a really interesting project for people looking for
dead zones (or illicit transmission) would be glasses that upconverted RF
into the visible spectrum, allowing you to look around and find the 'colour'
of wave you're looking for from different locations.

Matt MacLeod