Other Directory Sites: SeekWonder | Directory Owners Forum
The following archive was created by hippie-mail 7.98617-22 on Fri Aug 21 1998 - 17:16:22 ADT
Chuck McManis (cmcmanis@freegate.com)
Fri, 27 Mar 1998 17:23:01 -0800
Ok, so its becoming clearer, however I don't think one could reduce the
wheat/chaff packets to 1 bit. Given a 16 bit sequence and a 16 bit MAC
that would be a 33:1 wastage on good bits, plus n*33 bits for chaff
packets. So You're looking at turning a 1Mbit T-1 line into a 9600 baud
connection to achieve some privacy :-(. Also there is the question of
finding a MAC that functions on single bits, You see when there are only
two possible messages, there are only two possible valid MACs. My
intuition tells me that the length of the data in the message should
have at least as many bits as the MAC otherwise if you know the
algorithm you can easily compute which macs are good.
So for longish messages it gets exponentially harder, but can you crack
the MAC faster? This is another path I'll be looking at as well.
--Chuck
The following archive was created by hippie-mail 7.98617-22 on Fri Aug 21 1998 - 17:16:22 ADT