Re: Probably naive question - SHA1 + MD5 combination
- From: "Christian Siebert" <iBBiS@xxxxxx>
- Date: 31 Aug 2006 01:51:08 -0700
Multiple hash functions isn't a good idea.
What if we can find a set of hash functions that can be proven to be
independent? Can't we combine them to create a stronger hash function?
Simple example: Let's take the four hash functions {HAVAL-160,
RIPEMD-160, SHA-1 and Tiger(2)-160}, and let's assume that they are
independent (it's very likely that they are not!).
All of them take a message as input and produce a 160 bit message
digest as output. The output of the combined hash could be calculated
using XOR: 'C(M) = (H(M) + R(M) + S(M) + T(M)) mod 2^160'.
If the assumption would be true, then isn't breaking this combined hash
as difficult as breaking all of the used hashes separately? Provided
that at least 1 out of those 4 hashes remains invulnerable, the
combined hash should be invulnerable too. Or in other words: an
attacker needs to break all 4 hashes to break the combined hash.
Is this approach correct? If yes, how difficult would it be to prove
independence of hash functions?
Christian
.
- Follow-Ups:
- Re: Probably naive question - SHA1 + MD5 combination
- From: Kristian Gjøsteen
- Re: Probably naive question - SHA1 + MD5 combination
- References:
- Probably naive question - SHA1 + MD5 combination
- From: Shamus Husheer
- Re: Probably naive question - SHA1 + MD5 combination
- From: Kristian Gjøsteen
- Probably naive question - SHA1 + MD5 combination
- Prev by Date: Re: Probably naive question - SHA1 + MD5 combination
- Next by Date: Re: Probably naive question - SHA1 + MD5 combination
- Previous by thread: Re: Probably naive question - SHA1 + MD5 combination
- Next by thread: Re: Probably naive question - SHA1 + MD5 combination
- Index(es):
Relevant Pages
|
|