AN EFFICIENT TRUST MODEL FOR PUBLIC-KEY AUTHENTICATION AND DETECTING MALICIOUS NODES IN MANET

In Mobile Ad-hoc Networks (MANET), establishing trust among participating nodes is a challenging task due to the fact that these systems are generally operated in highly dynamic and distributed environments. A Pretty Good Privacy (PGP) like trust model considered to be more suitable for MANETs since they do not require access to a trusted authority. Instead, the nodes themselves are allowed to establish trust by certifying each other in a self-organizing way without involving any Trusted Third Party (TTP). Here, we adopt a strategy by using self-certifying ID-based cryptography that allows mobile nodes to determine public-keys of each other from the nodes' identities and their trust levels. And also, an Exclusion Access-control Mechanism based on trust metric evaluation is introduced to detect and isolate the malicious nodes that inject fake trust relations. We have proposed the certificate-less PGP trust model for public key authentication in MANETs. Hence, by improving trust metric in MANETs, trustworthy keys are determined through a given trust path. An exclusion mechanism is introduced to isolate the malicious nodes. Simulation results prove that the trust models and an exclusion access control mechanism provide an accurate malicious node detection and exclusion.