On Thu, 5 Dec 2002, Bon sy wrote: > I understand the inherent problem of RC4 is its structure on > exposing publicly expose the initialization vector and using finite number > of fixed shared "prviate" key, thus allowing --- at least in > theory --- sending a known simple text as a basis to decrypt some of the > messages when a key is reused. >
The problem is not with RC4, but how it's implemented in WEP. The 40-bit or the 104-bit "key" is part of the RC4 key, so certain attacks became feasible (especially when cards initialized the 24-bit IV to 0). The other issue was the known packet header structure of 802.11. I believe TKIP uses the key as a hash, so it's possible to use a bigger base of "keys" so that key reuse would not be a problem. Of course, I have yet to see firmware implementing it, although I've been told that it's possible to use TKIP in WEP-capable cards. Kevin "Starfox" Arima -- NYCwireless - http://www.nycwireless.net/ Un/Subscribe: http://lists.nycwireless.net/mailman/listinfo/nycwireless/ Archives: http://lists.nycwireless.net/pipermail/nycwireless/
