From: "Mullen Patrick" <Mullen.Patrick@mail.ndhm.gtegsc.com>
Date: Fri, 20 Sep 1996 02:32:59 +0800
To: "Cypherpunks" <cypherpunks@toad.com>
Subject: re: Stego inside encryption
Message-ID: <n1368993974.6219@mail.ndhm.gtegsc.com>
MIME-Version: 1.0
Content-Type: text/plain


To take this one step further, has anyone tried to ever use this method as
an encryption method?  You could hide data in a stream of random bits, using
position as the encryption method.  Obviously, the data would not be stored
in packets; rather as single bits strewn throughout the stream.  Even ASCII
characters could be hidden in such a system very well, as the possibility of
choosing the correct 8 bits (extended char set) from the data stream when
any combination has equal potential of being the correct sequence would be
extremely difficult.  Error checking/correcting code could even be used.

Using this system, the placement algorithm would be the focus of attack.  If
an algorithm which has a sufficiently random placement was used, extracting
the correct bits would be difficult.  Another way to increase the security
would be to hide the correct message inside a bitstream created by using
the same method on other similar messages.  (Hiding a real message inside
bogus messages.  Hmm...  Which one's real?)

Patrick

_______________________________________________________________________________
From: Mullen Patrick on Thu, Sep 19, 1996 9:19
Subject: Stego inside encryption
To: Cypherpunks

I know a lot of times the idea behind steganography is to hide the fact that a
secret message is in a seemingly normal file/mail/whatever.  This is good
for avoiding unwanted interest in your file.  The benefit of not having people
attempt to crack you code, added to the strength of the cryptosystem is
wonderful.  However, I propose this--

Don't hide that anything's encrypted!  Rather than hide this fact, throw it
in their face!  I propose hiding an encrypted message inside another 
encrypted message.  Set bits in specific places to data in the real message.
The benefit is Oscar not only doesn't know what the crypto is, he attacks
the wrong message.  Hiding statistically random bits from the true message
in statistically random bits from the masking message shouldn't be too 
hard.

Granted, this scheme doesn't get you past measures designed to keep out all
encrypted messages, and it surely wouldn't keep you message from generating
interest, but it would be very hard to decrypt the message, especially when 
some algorithm is used which (seemingly) randomly selects which bits to use 
for the stego.

Just a thought...  My apologies if someone has already proposed this method.

Patrick