[rc5] RC64 keyspace

Ray Vanlandingham jamesrv at awod.com
Tue Oct 28 06:21:34 EST 1997


At 11:09 AM 10/27/97 -0700, you wrote:
><snip discussion on OTPs in the digest...>
>>
>> What I was really wondering was whether we could use some analysis 
>>techniques to check high probability key space segments first, instead of 
>>just randomly checking the entire keyspace?  I read some information on 
>>this early in the 56bit days, but don't remember the specifics.  Has this 
>>already been ruled out?
>
>It is my understanding that the keyspace in the RC64 (or 56 or whatever)
>is basically totally flat.  That is to say, there is no section of the
>key space that has any higher probability of holding the solution than
>any other section of the keyspace.  That's what makes this form of
>encryption so valuable.  This is also why you cannot use A-life methods
>of cracking the encryption, there's no way to judge fitness.
>
>So the best way we have to break the encryption is to use brute-force.
>Unless you know something we all don't.  ;-)
>
>IM decided, I think, to avoid checking keys with duplicate numbers in
>them.  There's no statistical reason to do so and, in fact, it turned
>out they were incorrect in making this guess.

However if you get a 'partial' hit, say a key that decodes the 'T' at the
start of the message, could you use that information to start eliminating
keys? I not really that familiar with the encryption mechanism, but if part
of the key is used for each character, thenit such be possible to start
only sending out keys that match in that particular 'part' of the key. If
this would work, it could cut processing time immensely.


Ray Vanlandingham

----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list