Multivariate Fuzzy Analysis for Mobile Ad hoc Network Threat Detection

Detection of malicious, compromised and selfish node attacks is urgently needed to protect Mobile Ad hoc Networks (MANET) and their applications from failures. Once a MANET component is affected by a malicious, compromised or selfish node its operational state shifts from normal state to vulnerable state. Online monitoring mechanisms can collect important MANET network data that can be effectively used to detect abnormal behaviors caused by attacks. In this article we develop an online multivariate analysis algorithm to analyze the behavior of network resources and protocols to proactively detect MANET attacks. This algorithm is based on multivariate fuzzy analysis technique. We have validated the algorithm and demonstrated how it can effectively detect wellknown attacks such as Denial of Service in MANET.