Re: hash function
From: Aldar C-F. Chan (aldar_at_comm.utoronto.ca)
Date: 02/21/05
- Next message: Jim Gillogly: "Re: A challenge...!"
- Previous message: Chris Shearer Cooper: "Crypto++ really slow why?"
- In reply to: David Wagner: "Re: hash function"
- Next in thread: David Wagner: "Re: hash function"
- Reply: David Wagner: "Re: hash function"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Mon, 21 Feb 2005 22:03:09 GMT
> In effect, you want F_k(x) = hash(x||a) to be a PRF. Any chance you can
> use a standard PRF instead of a hash-based construction? For instance,
> something like AES-OMAC, SHA1-HMAC, etc.?
I need the hash function be public, so PRF does not fit. If I relax the
requirement
to just keeping x secret (instead of guarding against a collision to x), I
am
wondering if hash(x||a) = E_{f(x)}(a) works (with f(x) being another hash
function).
- Next message: Jim Gillogly: "Re: A challenge...!"
- Previous message: Chris Shearer Cooper: "Crypto++ really slow why?"
- In reply to: David Wagner: "Re: hash function"
- Next in thread: David Wagner: "Re: hash function"
- Reply: David Wagner: "Re: hash function"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]