Markov Chain-Based Trust Model for Analyzing Trust Value in Distributed Multicasting Mobile Ad Hoc Networks

The multicast mobile ad-hoc network (multicast MANET) is proposed for a sender sending packets to several receivers through a multicast session. Since members of a multicast group changes frequently, the issue of supporting secure authentication and authorization in multicast MANET becomes more critical than the network provided a fixed central authentication (CA) server. This work thus proposes a two-step secure authentication approach for the multicast MANET. First, a Markov chain trust model is proposed to determine the trust value (TV) for each one-hop neighbors. A nodepsilas trust value is analyzed from its previous trust manner in this group. The proposed trust model is proven as an ergodic continuous-time Markov chain model. Second, a node with the highest TV of a group will be selected as the CA server. For increasing reliability, the node with the second highest TV will be selected as the backup CA server that will take over CA when CA fails. The procedures of the secure authentication for group management are detailed. The security analysis of each procedure is analyzed to guarantee that the proposed approach achieves a secure reliable authentication in multicast MANET. Numerical results indicate that the analytical trust value of each mobile node is very close to that of simulation under various situations. The speed of the convergence of the analytical trust value shows that the analytical results are independent of initial values and the trust classes.