On Thu, Mar 13, 2003 at 03:54:51PM -0600, Dan Minette wrote:

> In some cases, we are actually close to it.  The station that
> broadcasts the Houston Rockets have change their signal pattern to not
> interfere with a station up in Kansas after 6PM. So, I have a very
> hard time hearing them over other stations.
>
> Radio receivers have traditionally operated on the principal of a
> tunable resonance.  Such a resonance has finite width, so one cannot
> have a radio station at 601.1 kHz and another at 601.2 kHz and expect
> to receive one and not the other.  That is why there is "only so much
> room on the dial."

But is the "room on the dial" used efficiently? Would the Houston/Kansas
radio problem still exist if things were organized differently?

Since the radio band is presently partitioned into fixed channels, you
may have some channels that are being used, and others which are not
being used. Also, most audio streams can be digitally compressed on
the fly by 5 to 10 times.  So, for example, if we were to switch to
ultrawideband (UWB) and pseudorandom noise (PN) coding, and add some
sort of protocol for the compression, and implement a channel code
lookup system, then bandwidth could be more efficiently utilized.

It is also worth noting that if you went even further and divided
space into cells, with adjacent cells having different multiplexing to
distinguish them, then the physical limit will only apply within each
cell, so you could multiply your overall capacity to some extent by
going to a cellular system. Using a packetized data system (like the
Internet) you could route a signal over one spatial path, and another
signal over a different spatial path. But you can only make the cells so
small (it wouldn't make sense to have cells smaller than the physical
size of the equipment transmitting and receiving signals), so there is
still a limit.

The limit (for each cell) is defined in Shannon's theorem:

Channel Capacity (bits/sec) = Bandwidth * log_2 ( 1 + S/N )

Here's a decent discussion of RF communications and bandwidth
utilization (note the chart that gives the fraction of Shannon's
theoretical limit reached by various coding schemes):

  http://tinyurl.com/7gn1


-- 
"Erik Reuter" <[EMAIL PROTECTED]>       http://www.erikreuter.net/
_______________________________________________
http://www.mccmedia.com/mailman/listinfo/brin-l

Reply via email to