RSS-based spoofing detection and localization algorithm in IEEE 802.11 wireless networks

Wireless networks have gained immense popularity in recent years. Though wireless networks have innumerous advantages over conventional wired networks, the borderless nature of wireless networks makes it prone to various felonious activities. In the context of network security, one of the most crucial vulnerabilities which have caused a serious global concern is the MAC Address Spoofing. MAC address spoofing facilitates launching of other attacks such as Denial-of-service, Man-in-the-middle, SYN flooding etc. In this paper we propose a methodology for the detection of spoofing attacks in IEEE 802.11 networks that involve performance of cluster analysis on RSS patterns of 802.11 transmitters which not only detect presence of a spoofing attack but also determine number of attackers. Further for localizing the adversaries, we propose a discriminant-adaptive neural network (DANN) based localization system.