1993-03-11 - FW: Hiding Encrypted Messages

Header Data

From: Stu Klingman <stuk@microsoft.com>
To: cypherpunks@toad.com
Message Hash: 11f99c3c21d6b85d7bf76f47cfe82db77a07e482be9496fcfba6239a86b8abb4
Message ID: <9303112315.AA12701@netmail.microsoft.com>
Reply To: N/A
UTC Datetime: 1993-03-11 23:18:49 UTC
Raw Date: Thu, 11 Mar 93 15:18:49 PST

Raw message

From: Stu Klingman <stuk@microsoft.com>
Date: Thu, 11 Mar 93 15:18:49 PST
To: cypherpunks@toad.com
Subject: FW: Hiding Encrypted Messages
Message-ID: <9303112315.AA12701@netmail.microsoft.com>
MIME-Version: 1.0
Content-Type: text/plain

FWD'd from a secure redirector, Raf sez:

Someone wrote:
> b) If it's played and recognized - one can trace your source (a CD, a
>    tape of radio broadcast, whatever) and do a comparison. Then the
>    file containing of all the LSBs is cryptanalyzed...

Actually, this is not really a problem.  The odds of being able to
resample, even using the same source and come up with the same byte
string is infinitesmal.  You've got chaos theory on your side here
with massive "Sensitive Dependence on Initial Conditions."  The
exact byte string you got depends on:

1) volume
2) sampling device used
3) playback fidelity of your reproduction
4) the exact microsecond you clicked "record" (cause you'll be hitting
   different points in the same wave form)

Just make sure to an application like Shredder or Flamefile to
permanently erase your initial sample, and nobody should be able
to tell. (unless they are aware of the trick beforehand)