REPSYS: A Robust and Distributed Reputation System for Delay-Tolerant Networks

Distributed reputation systems can be used to foster cooperation between nodes in decentralized and self-managed systems due to the nonexistence of a central entity. In this paper, a Robust and Distributed Reputation System for Delay-Tolerant Networks (REPSYS) is proposed. REPSYS is robust because despite taking into account first-and second-hand information, it is resilient against false accusations and praise, and distributed, as the decision to interact with another node depends entirely on each node. Simulation results show that the system is capable, while evaluating each node's participation in the network, to detect on the fly nodes that do not accept messages from other nodes and that disseminate false information even while colluding with others, and while evaluating how honest is each node in the reputation system, to classify correctly nodes in most cases.