Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!ll-xn!cit-vax!tybalt.caltech.edu!palmer
From: palmer@tybalt.caltech.edu (David Palmer)
Newsgroups: sci.math,sci.crypt
Subject: Encryption of Hoffman coded text
Message-ID: <3515@cit-vax.Caltech.Edu>
Date: Thu, 6-Aug-87 20:35:17 EDT
Article-I.D.: cit-vax.3515
Posted: Thu Aug  6 20:35:17 1987
Date-Received: Sat, 8-Aug-87 14:32:59 EDT
References: <10489@linus.UUCP] <7876@mimsy.UUCP> <10604@linus.UUCP>
Sender: news@cit-vax.Caltech.Edu
Reply-To: palmer@tybalt.caltech.edu.UUCP (David Palmer)
Organization: California Institute of Technology
Lines: 23
Xref: mnetor sci.math:1771 sci.crypt:515

If you are going to encrypt a message by the standard breakable techniques,
(Caesar cypher, Ultra, Purple, NBS DES, whatever technique the NSA is
selling to our allies :-)  Is decryption more difficult if you use
Hoffman coding (including Hoffman coding of common combinations such as 'th')
on the plaintext before encypherment?  I am assuming that the people doing
the attack know that you are doing Hoffman coding and know what the code is.

(For those who are not familiar with the technique, Hoffman coding uses
variable length characters, the more common characters requiring fewer
bits.  This usually decreases the total length of a message (in bits), assuming
that it is non-pathological and not overly concerned with zebu and oxen in
Zanzibar and Qom)

It seems to me that Hoffman coding makes the plaintext look more 'random'
and, so, less amenable to statistical attack.  Is this so?

I don't know how to crack anything harder than a simple substitution
cypher, so details would be welcome.


		David Palmer
		palmer@tybalt.caltech.edu
		...rutgers!cit-vax!tybalt.caltech.edu!palmer