Cluster-Based Trust Model against Attacks in Ad-Hoc Networks

Abstract

In ad-hoc environments, all nodes have function to set themselves a route and build a network. It is important how trustable each node is in the environment. This study proposes the cluster-based trust model, and uses the model to discuss how you can solve some possible problems occurred in the network. The reputation from a neighboring node is applied to the calculation of the trust value. If the trust value of a truster is used as a weight, more sophisticated calculation of the trust value is available. When you create a cluster, all nodes in the cluster can fully trust the selected head, if entire nodes in the cluster participate in the head competition. And then, the head node issues the certificate that shows the trust level of each member node. If a node moves from one cluster to another, the trust level of the node is determined by the certificate issued by the previous cluster-head. The model proposed by this study can monitor the selfish behavior of nodes in the network and isolate the selfish nodes from the network to enhance the effectiveness throughout the network.