James Maitland (jabba@jcp.co.uk)
Fri, 04 Sep 1998 09:09:10 +0100
Hiya,
I'm getting a bit confused by the different descriptions of the
blum-blum-shub PRNG- one from Stinson, the other Schneier. I've no
problem with the generation of the large primes (p and q) used to
generate the modulus (n). But the seed (s0) is proving troublesome- or
rather, I suspect I've got it wrong.
Stinson describes how the seed, s0 is an element of the quadratic
residue set QR(n). Since n is the product of the two primes p and q, the
membership test involves calculating the jacobi symbols for:
( s0 / p ) and
( s0 / q )
and checking these both equal +1
Okay, that's cool. The question is:
If I generate some large random number (a), and it passes these jacobi
tests, is that my value for s0?
Or should s0 equal [ a^2 mod n ]? Schneier's description doesn't mention
Jacobi symbols, simply saying to choose an 'X' that is relatively prime
to n, then set s0 to [ x^2 mod n ].
Any clarification/explanation would be most welcome.
merci,
jabba.
-- James Maitland Senior DeveloperJCP Computer Service Tel: +44 (0) 171 689 6890 16 St Johns Lane, Fax: +44 (0) 171 689 0438 London, E-Mail: jabba@jcp.co.uk EC1M 4BS WWW: http://www.jcp.co.uk United Kingdom
-- PGP Public Key http://www2.jcp.co.uk/keys/http://www2.jcp.co.uk/keys/james@jcp.co.uk.txt
The following archive was created by hippie-mail 7.98617-22 on Sat Apr 10 1999 - 01:13:58