From: ccat@netcom.com (Chris Beaumont)
Date: Sun, 12 Sep 93 09:14:18 PDT
To: cypherpunks@toad.com
Subject: encryption key switching,would it work?
Message-ID: <9309121608.AA08172@netcom5.netcom.com>
MIME-Version: 1.0
Content-Type: text/plain


I'm wondering if there is any particular reason why an encrypted message
might not be encrypted with several different keys,within the same message,
divided by time.Perhaps synchronized clocks or time stamping might be used
to toggle the encryption parameters.This technology might be able to defeat
the brute force decryption technologies described in the paper mentioned
yesterday.
The only way that one might decrypt the message is to know the _exact_ time
at wich the algorhythm was switched.In other words, what to prevent people
from using _numerous_ keys during _one_ session,with the _times_ of
switching remaining secret?
Combined with a robust "one time pad" method,like triple DES,this
technology might be particularly resistant to casual decryption attempts
based on knowledge of the algorhythm? I know little about different
encryption methods.. is there
any reason this wouldnt work,besides the necessary complexity of the
information contained in the key?
-Chris
-
Chris Beaumont                    nutrient cafe wholesale
ccat@netcom.com                  MIME mail graciously accepted
ccat@casa.stanford.edu           public key available via finger