1998-12-04 - Re: Streams, Voice, and Sensitive Dependence

Header Data

From: Jim Gillogly <jim@acm.org>
To: cypherpunks@cyberpass.net
Message Hash: c912471fc85670c33d86ab741395327055e929e2d691867ffd6952f7f5a9ab09
Message ID: <36683D66.CCBCB35F@acm.org>
Reply To: N/A
UTC Datetime: 1998-12-04 20:23:49 UTC
Raw Date: Sat, 5 Dec 1998 04:23:49 +0800

Raw message

From: Jim Gillogly <jim@acm.org>
Date: Sat, 5 Dec 1998 04:23:49 +0800
To: cypherpunks@cyberpass.net
Subject: Re: Streams, Voice, and Sensitive Dependence
Message-ID: <36683D66.CCBCB35F@acm.org>
MIME-Version: 1.0
Content-Type: text/plain

"Lee Davis"<lee.davis@thomasmore.edu> skribis:
> In any case, there has been a lot of recent work in dynamic systems (Chaos),
> especially in the fast computations of Julia Sets.  Has anyone seen a paper on
> exploiting the sensitive dependance in these systems for pseudorandom number
> generation?

A number of attempts have been made to apply chaos to cryptosystems.  The two
that I've broken suffered from the same fault: although it's difficult to find
the precise starting point (key) due to sensitive dependence, chaotic systems
are by definition non-random, and have preferred orbits.  If the cryptanalyst
finds an orbit that's close to the one used by the actual key, the stream is
mostly the same; this is good enough to hone the attack for the next pass.

I'm not saying chaos has no applications in cryptography; only that the
applications are not obvious.
	Jim Gillogly
	Trewesday, 14 Foreyule S.R. 1998, 19:37, 6 Manik 20 Ceh, Sixth Lord of Night