A Collaborative Reputation Mechanism for Underwater Acoustic Networks

Distributed trust mechanisms can be used to protect Underwater Acoustic Networks (UANs) against a variety of Denial of Service (DoS) attacks. Several UAN trust systems have been proposed in the recent years, but only a few of them exploit Trust-Related Data (TRD) dissemination to improve the knowledge that a node has about its neighbors. Without TRD sharing, nodes can only form an opinion about their 1-hop neighbors, which might also be inaccurate due to packet losses being mistaken for actual malicious behavior. This paper presents a collaborative reputation algorithm for UANs that performs reputation sharing, propagation and aggregation allowing distant nodes in the network to form an opinion about each other and increasing the overall reliability of the trust metric. The proposed security mechanism is then implemented and tested as a DESERT Underwater module to verify its quality. Results show that trust values resulting from the aggregation of second-hand opinions are more reliable than the first-hand evaluations obtained through direct experience, effectively extending the knowledge that each node has about the trustworthiness of the others.