Chaotic encryption research being evaluated

[Mads Rasmussen]

Giving a recent thread on sci.crypt about chaotic encryption I thought
this seemed interesting
 
The actual paper is available from the link below
 
>From Security Wire Digest Vol 5 No 5
 
The 
 
---
*CHAOTIC ENCRYPTION RESEARCH BEING EVALUATED
By Carl Weinschenk
A new approach to chaotic encryption proposed by researchers at Beijing
Normal University has a good pedigree, but many experts are taking a
wait-and-see attitude on its speed and security.
 
Chaotic encryption employs an algorithm to transmit information in a
stream that grows increasingly disordered--or chaotic--over time. The
receiver can use the algorithm to remove the random data. This, in
essence, rolls back the clock to reveal the initial data.
 
The challenge to the commercialization of any new encryption
technique--one that chaotic encryption hasn't reached--is proving it's
as fast and secure as existing systems, such as TripleDES, says Jon
Callas, CTO of PGP Corp.
 
The research on encrypting two-way voice communications using chaotic
encryption was reported in Physical Review E. Dr. Hu Gang, who led the
research, says that the approach has commercial potential because it
uses fast "single round" analytical computations. The speed is achieved
without compromising security because the type of chaos
used--"spatiotemporal"--is highly secure. "The particular advantage with
our system is that we can produce ciphertexts in each round," he says.
 
"Professor Hu...has a very talented group of researchers and does
excellent work," says Rajarshi Roy, a professor in the Department of
Physics and the Institute for Physical Science and Technology at the
University of Maryland. "As a method of speech encoding using computer
hardware and software, this is an interesting technique, and its privacy
and security need to be studied."
 
But Janusz Szczepanski, a researcher at the Polish Academy of Sciences'
Institute of Fundamental Technological Research, says the authors have
only considered one type of attack, so the jury is still out on the
quality of this method's security.
 
"I have looked at their system and, without detailed analysis, I can
point out some points that can be attacked," says Szczepanski. He awaits
the further research that the paper's authors say is coming.
http://ojps.aip.org/getabs/servlet/GetabsServlet?prog=normal
<http://ojps.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=PLEEE80
00066000006065202000001&idtype=cvips&gifs=yes&jsessionid=339324104281771
0305>
&id=PLEEE8000066000006065202000001&idtype=cvips&gifs=yes&jsessionid=3393
241042817710305
 
---
 
Mads Rasmussen
Open Communications Security
+55(11)3345-2525
 

---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at wasabisystems.com