Abstract
This paper presents a genetic programming approach to
detect deauthentication attacks on wireless networks
based on the 802.11 protocol. To do so we focus on
developing an appropriate fitness function and feature
set. Results show that the intrusion system developed
not only performs incredibly well - 100 percent
detection rate and 0.5 percent false positive rate -
but also developed a solution that is general enough to
detect similar attacks, such as disassociation attacks,
that were not present in the training data.
Users
Please
log in to take part in the discussion (add own reviews or comments).