On improving the efficiency of intrusion detection systems using game theoretic approaches
No Thumbnail Available
Date
2018
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The thesis consists of three distinct contributions. As the first contribution, a novel game theory-based false alarm minimization scheme for signature based IDS is proposed. The proposed framework models the intrusion detection process as a two player noncooperative game between the IDS and the attacker. It uses various network context information like, IDS’s detection rate, criticality levels of the host machines, severity levels of network vulnerabilities, attacking and monitoring costs etc., to devise efficient IDS monitoring strategies based on the Nash Equilibrium (NE) of the game. The proposed framework is shown to filter out most of the false positive alarms generated by the signature based IDS and thereby, significantly improve the IDS’s accuracy, without adversely affecting its detection capabilities
Description
Supervisors: Sushanta karmakar & Santosh Biswas
Keywords
COMPUTER SCIENCE AND ENGINEERING