Abstract: Trust and reputation models are utilized in the security mechanisms of MANET to deal with selfish and misbehaving nodes and ensure safe delivery of packets from source to destination. The calculation of trust values and its propagation between the nodes form the rubric of these trust models. But its effectiveness is often degraded by the propagation of fake trust values via. dishonest recommendations particularly in environments like MANET due to the lack of centralized administration and mobility of nodes. However, dealing with dishonest recommendation attacks is a highly challenging task. We propose an enhanced integrated dynamic trust recommender that filters out dishonest recommendations by clustering, filtering and selecting the recommended trust values based on certain criteria. This makes the trust reputation model more robust and accurate in the dynamically changing MANET environment.
Ramesh Ponnusamy, H. Abdul Raufh and S. Malarvizhi, 2017. Detection of Malicious Nodes using Enhanced Integrated Dynamic Trust Recommender in MANET. Asian Journal of Information Technology, 16: 181-189.