An approach to detect malicious node for Delay Tolerant Networks

Abstract

In Delay Tolerant Networks (DTNs), the information is transferred from its source to destination without end-to-end connectivity of the network. The Reputation-Based Trust Management Systems for MANETs do not apply to DTNs due to lack of end-to-end connectivity. The existing Iterative Trust Reputation Mechanism (ITRM) is effective in a detection of malicious node, however it faces from latency time, and hence, sometimes results in false detection of node considered as malicious node. The Proposed Modified Iterative algorithm for Detection of the malicious node for Delay-Tolerant Networks is based on clustering of Raters as per their ratings to Service Providers. The Raters are categorized as low priority, middle priority and high priority based on rating given to Service Provider. This has resulted into lowered amount of false positives and improvement in Service Provider reputation accuracy. The Proposed Modified Iterative algorithm for Detection of the malicious node for Delay-Tolerant Networks is compared with the existing ITRM algorithm and found that, the proposed algorithm has less time complexity and better accuracy of detection of malicious nodes.