Re: jointly create a random value for corrupted party
From: Anna Rikova (nospam_at_nospam.com)
Date: Sun, 17 Jul 2005 21:13:14 +0200
> (a) Does B know the values that C and D sent to A? (What
> happens if A tells B the values that were sent by C and/or
B, C and A work togehter. They are all corrupted by an adversary A. So B
and C knows everything what A knows and vice-a-vers. The honest party D
only knows what is publicly known.
> (b) Can B benefit by sending a "non-random" value to A?
> (How?) What if B and C conspire? What if B, C, and D
It is assumed that always one party is honest. But not the party for
which the random value is created.
> (c) Can A benefit by "cheating"; can he report that he is
> using one random number while actually using a different
> random number? (For what purpose is A using the random
He uses it as part of a key. He can benefit in using a non-random value,
cause maybe then he could "choose" a weak key, or influence the keys
this way, that the group to seach for the key is smaller.
> (d) Should the random number used by A be kept secret from
> B, C, and D?
> It might be helpful for you to describe the "real-world"
> problem you are trying to solve. You may have too many (or
> not enough) constraints in your abstract description.
The "real-world" problem is, that A should be forced to use a random
value as a key.
Thanks in advance,