From: Bill Stewart <stewarts@ix.netcom.com>
Date: Wed, 13 Dec 1995 05:29:44 +0800
To: cypherpunks@toad.com
Subject: Re: chaotic signal cryptography
Message-ID: <199512122008.MAA09312@ix11.ix.netcom.com>
MIME-Version: 1.0
Content-Type: text/plain


At 08:00 PM 12/12/95 +0100, Anonymous wrote:
>>A few years back i ran thru a paper emanating
>>from a navy labs researcher (i think the name was Peccora)
>>about using synchronised chaotic signal generators as a
>>signal scrambling/encrytpion scheme.
>
>I too am interested in this technique.  I may be wrong, but it seems like
>a variation on the one-time pad.  Any comments on this technique (usefulness,
>robustness, etc.)?

1) It's not, repeat NOT, a one-time pad, or anything like one.
Synchronised generators on both ends of a communication are yet another
keyed algorithmic system, and are theoretically breakable; 
one-time pads are theoretically unbreakable (as long as you only use them
once, prevent theft, etc.).  Whether algorithmic systems are breakable 
in practice depends on how good the algorithms are, how good the keys are,
how much attention is paid to cracking them, etc.

2) Almost everything I've heard about chaotic cryptosystems has said
they're not very effective; it's too easy to predict the output.
They're fine random number generators for simulations, but they're
not cryptographically strong.

3) Maybe somebody's done something new and interesting in the field
and hasn't told the general public about it, but if you heard about it
a few years back it's been broken.
#--
#				Thanks;  Bill
# Bill Stewart, Freelance Information Architect, stewarts@ix.netcom.com
# Phone +1-510-247-0663 Pager/Voicemail 1-408-787-1281