privacy amplification

From: Chris Muktar (chrismuktar_at_gmail.com)
Date: 11/26/04 

Date: 26 Nov 2004 03:04:16 -0800

Hi,

I'm just trying to understand "Privacy Amplification" and am working
from paper [1].

I understand that Bob and Alice randomly select a compression function
g from a set of compression functions G, and encode the secret K =
g(W) where W is the input string.

It is understood that Eve has full knowledge of the set G, and
therefore also g, but it is also understood that if Eve does not know
which g to select, she will be unable to deduce the secret K from the
input string W.

The problem I encounter is, how do Bob and Alice agree which g they
are going to use, such that their compression functions coincide?

Note: I am working in the context of quantum cryptography, so that
P[VW] is specific to that application.

Cheers!
Chris

[1] "Generalized Privacy Amplification", Bennet, Brassard, Crepeau,
Maurer, IEEE Transactions on Information Theory Vol 41, No 6, November
1995

Relevant Pages

  • Re: privacy amplification
    ... Eve does not have the same identical string as Bob and Alice, ... such that their compression functions coincide? ...
    (sci.crypt)
  • Re: privacy amplification
    ... > I understand that Bob and Alice randomly select a compression function ... she will be unable to deduce the secret K from the ... such that their compression functions coincide? ... You say you're working in the context of quantum cryptography which ...
    (sci.crypt)