A Reputation-Based Trust Model for P2P Systems

One of the fundamental challenges for Peer to Peer (P2P) systems is the ability to manage risks involved in interacting and collaborating with prior unknown and potentially malicious parties. Reputation-based trust management systems have proven to be essential to enforcing cooperative behavior in P2P networks. However, in many existing reputation-based trust models, the validity of peers´ trust valuation is seriously affected by peers´ malicious behaviors, such as strategic cheating, dishonest recommending and collusion. In this paper, we propose a novel reputation-based trust model for P2P systems. In our approach, the trustworthiness of peers consists of the experience-based and recommendation-based trust. To provide adequate reaction to peers´ malicious behaviors, some factors are introduced to distinguish these experiences about peers´ behaviors. Simulation shows that the proposed model has advantages in countering peers´ malicious behaviors over the existing trust models.