home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!sun-barr!rutgers!igor.rutgers.edu!zodiac.rutgers.edu!leichter
- From: leichter@zodiac.rutgers.edu
- Newsgroups: sci.crypt
- Subject: Re: Triple DES
- Message-ID: <1992Nov20.094321.1@zodiac.rutgers.edu>
- Date: 20 Nov 92 14:43:21 GMT
- References: <921116133628.385022@DOCKMASTER.NCSC.MIL> <1eg516INNrrq@uniwa.uwa.edu.au>
- Sender: news@igor.rutgers.edu
- Organization: Rutgers University Department of Computer Science
- Lines: 41
- Nntp-Posting-Host: pisces.rutgers.edu
-
- In article <1eg516INNrrq@uniwa.uwa.edu.au>, markd@uniwa.uwa.edu.au (Mark
- Diamond) writes:
- | The question arose in a previous reply to this posting of whether
- | triple, double or any other repeated DES encryption was stronger than a
- | single encryption. The previous reply indicated that the question was
- | still open. It is not open ... it has been settled and the results
- | presented at CRYPTO 92.
- |
- | Essentially the question about the strength of double encryption is the
- | same as the question about whether DES is or is not a group.
- |
- | If DES were a group then for two keys K1 and K2
- |
- | DES( DES(Plaintext, K1), K2) == DES(Plaintext, K3) for some
- |
- | K3 which presumably the eavesdropper is trying to discover ... the point
- | being that they would not have to find K1 and K2, but only a kind of
- | "combination" K3.
- |
- | The recently demonstrated fact that DES is *not* a group indicates that
- | a double encryption *is* stronger than single encryption.
-
- Sorry, no. Repeating a falsehood does not make it true. All we know is that
- it MIGHT be stronger.
-
- Consider TES, the Trivial Encryption Standard - it's just like DES except that
- TES(P,k) = k || DES(P,k) (|| is concatenation). That is, TES "gives away the
- store" at the beginning of the ciphertext. It's clear that if DES isn't a
- group, neither is TES. It's also clear that double TES is exactly as strong
- as single TES.
-
- If there were a generic unknown-plaintext attack against DES, it would work
- just as well against double DES, requiring double the time. (Yes, this would
- make double DES just a bit stronger - but multiplicative factors - especially
- such small ones! - are ignored when we discuss the strength of cryptosystems.
-
- A pure chosen-plaintext attack against DES would not move directly to double
- DES. But that only says something about ONE PARTICULAR ATTACK; it says
- nothing about the inherent strength of the cryptosystem.
-
- -- Jerry
-