Tracy writes:

> We've been out-foxed, time to give up on the SETI project:
>
>  http://www.spacedaily.com/news/seti-04k.html

I've already been asked about this twice now before. Here's what I wrote in
my first reply:

=======================================

> What do you think of this?  How could a blackbody radiation curve carry any
> information unless it was modulated somehow?  Just wondered if you had run
> across this idea.

Several things seem to have gotten conflated in all of this. It is absolutely
true that the most compressed signal possible is indistinguishable from
random noise. It is a signal in which all redundancy has been removed, so that
nothing is predictable from one transmitted symbol to the next.

Unfortunately, such a signal is also the most fragile and one that can't be
transmitted asynchronously. Some form of regular "keying" is necessary so that
a remote receiver can lock onto it, and that alone would make it
differentiatible from random noise. Secondly, all information recovery techniques that are
imbedded in a signal impose some sort of redundancy into the signal. Without
that redundancy, the information contained in a corrupted packet could not be
recovered, and that redundancy too would make the signal differentiatible.
(Without information-recovery capabilities, a corrupted packet would need to be
asked to be retransmitted, and given the penalties of the speed of light, no one
would design a long-distance communications network that way).

Thirdly, although I didn't see how in the text the two were philosophically
equated, a random source need not be a blackbody source. Blackbody is for
certain a random assemblage of events, but the inverse isn't true.

Fourthly, spread spectrum isn't random. The idea was invented during WWII as
a method to scramble radio communications so as to make the message
"untappable". The notion was simple enough, and a piano keyboard was the inspiration.
Transmit a part of the message on one key, then shift to another preplanned key,
transmit a part of it there, and then shift again, and so on. Unless you knew
the sequence of frequency hops in advance, you couldn't begin to hear much
more than a few snippets of the conversation.

The inventor of spread-spectrum wasn't just another pretty face, btw. It was
Hedy Lamarr, the actress, and she's received IEEE's highest honors. See:

http://www.spectrum.ieee.org/careers/careersarchive/careerstrategy/aug04/0804c
aref1.html

========================================

> Thanks for the very interesting response.  I only thought of blackbody
>  radiation because they said it would be indistinguishable from star light,
>  which is pretty close to blackbody radiation if I understand correctly.

If viewed over a fairly restricted range of frequencies ("one color" or so),
the radiation from a star is indistinguishable from a random signal. However,
when sampled over the broad range of frequencies that a star produces, the
differences would be quite obvious. There would be no reason to transmit over any
broader range of frequencies than that required by the modulation bandwidth
of the signal you're transmitting. Anything else would be an enormous waste of
power and no civilization worth its salt would do such a thing.

========================================

Wirt Atmar

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *