What the heck is this? Optical noise encryption?

Anyone know anything about this?  How secure is this? If you have to have
2 "identical" lasers to pull this off, sounds like this would not be very
secure, since there must be a pretty wide margin of error (I mean, how
"identical" can 2 lasers be?)


____ from EduPage ____

OPEN SECRET
    Instead of using mathematical codes to scramble and unscramble
messages, Georgia Tech physicists are devising a way of sending a message
with electronic noise generated by a flickering laser. By connecting
identical lasers over fiber optics, the same random pattern of noise is
generated at both the sending and receiving end, and the receiving simply
subtracts the noise to uncover the message. (Atlanta Journal-Constitution
4/7/94 E2)
__ end _______________

 

-- 
Stanton McCandlish * mech@eff.org * Electronic Frontier Found. OnlineActivist
"In a Time/CNN poll of 1,000 Americans conducted last week by Yankelovich
Partners, two-thirds said it was more important to protect the privacy of
phone calls than to preserve the ability of police to conduct wiretaps.
When informed about the Clipper Chip, 80% said they opposed it."
- Philip Elmer-Dewitt, "Who Should Keep the Keys", TIME, Mar. 14 1994

I don't know much about his "optical noise" encryption, but Scientific
American ran an article in the Amatuer Scientist column outlining an
encryption scheme involved much the same ideas.  Their scheme was to
find, and digitize a chaotic source as a carrier signal and then add the
information to the signal. Decryption involved subracting the source
signal from the encoded one.   Information encrypted this way would be
positively unbreakable by anyone without the chaotic source, as the
whole scheme is entirely random. 

The biggest problem is of course, transmitting the source in a secure
manner.  It is not enough to know the configuration of the generator of
the source, unless you also the _exact_ initial conditions (which are,
of course, impossible).

One solution to this problem might be to use a recursive equation to
generate a source from a small (one-hundred+ digit)  seed, and the
number of iterations necessary to reproduce the source.  The nice thing
about this is that the equation could also be customized, something 
like: x= (k)(x^2)+a,  where "a" and "k" are constants that may be
altered, thus providing two methods of encryption. This scheme would
also be a solution to the problem of the source being corrupted during
transmission (which would ruin any attempts to use it).  But the
equation and the seed would still have to be transmitted somehow.
Hmmm.... you could openly send the seed and  the encoded information,
and then call the reciever to tell them to convert a given sentence into
decimal equivalent...etc

Ethan D Schartman says:
> One solution to this problem might be to use a recursive equation to
> generate a source from a small (one-hundred+ digit)  seed, and the
> number of iterations necessary to reproduce the source.  The nice thing
> about this is that the equation could also be customized, something 
> like: x= (k)(x^2)+a,  where "a" and "k" are constants that may be

And at this point I've just built an ordinary PRNG+Xor based stream
cipher and the usual techniques to break it all apply.

All this "encryption with chaos" stuff just adds up to "look at my
fancy PRNG", which of course is a game that crypto people have been
playing for a long time. I'm starting to get alarm bells go off every
time "chaos" is mentioned.

Perry