[12023] in cryptography@c2.net mail archive
Re: DOS attack on WPA 802.11?
daemon@ATHENA.MIT.EDU (William Arbaugh)
Fri Nov 8 09:55:19 2002
Date: Fri, 8 Nov 2002 09:40:50 -0500
Cc: "Arnold G. Reinhold" <reinhold@world.std.com>,
cryptography@wasabisystems.com
To: Niels Ferguson <niels@ferguson.net>
From: William Arbaugh <waa@cs.umd.edu>
In-Reply-To: <3.0.3.32.20021108030722.0091a7a0@pop.xs4all.nl>
TGi has NEVER been all that interested in DOS attacks because a number
of people argued that all you need to do is turn on a spark gap
transmitter. While this is true, I think it is harder (one can argue
how much) to get a spark gap transmitter and use it correctly than a
laptop, NIC card, and parabolic dish. As a result, the threat class
becomes much larger than it should be. And BTW, you can do all sorts of
DOS attacks against the base .11 protocol (sending management, EAP,
etc. frames willy nilly; see http://802.11ninja.net/ as an example).
I think the bigger concern with the Michael countermeasures is:
1. Will the vendors implement them, and
2. Will they be implemented correctly?
Ideally, the compliance checking will ensure this.......but then
again......
TGi had do a delicate balancing act between finding a solution that can
be implemented in firmware, and actually makes some improvements. I
think they did a reasonable job with WPA1 considering the engineering
challenges.
On Thursday, Nov 7, 2002, at 21:07 US/Eastern, Niels Ferguson wrote:
> Yes, the Michael countermeasures allow a DOS attack. This was widely
> discussed in 802.11-TGi before the countermeasures were accepted.
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo@wasabisystems.com
