Research On Trust Model For P2P Networks

With the rapid development and popularization of Internet the P2P network technology has gradually become a focus of research and concern. It is widely applied in the fields of file-sharing, distributed computation, cooperation system, deep search engine and electrical business, and has a brilliant future. Due to the P2P network's opening and anonymity, there exist more and more misbehaviors of malicious peers severely endangering P2P services. Besides, there are always a large numbers of "Free-Riding" users in the network, which consume resource but do not make any contribution. The selfish behaviors cause the degradation of network performance. Therefore how to identify the malicious peers, restrain the misbehaviors of dishonest peers, incent peers cooperate and consume resources rationally have become important in the trust mechanism.Based on the research of typical resolve of "trust mechanism" and "incentive mechanism", the thesis proposes a recommendation-based trust model with reference to the trust relationship of communication network of human society and a "different service"-based incentive mechanism which is fit for the trust model. The model evaluates trust by estimation..The trust are divided into the direct trust and recommendation trust by the collection and analyzing and refine the recommendation trust for friend trust and stranger trust The reputation value of peer is calculated by the three as the proof of trade. The incentive mechanism described adopts "different service" idea and the standard comes from the two measurement for incite: global reputation contribution and global flow contribution ,the former is deemed as measurement for quality of file sharing service and the latter represents the contribution of node to upload flow of all nodes. The simulation results proved that the trust model has better performance in evaluation accuracy, security, practicability, resistance to attacks on the malicious peers and the incentive mechanism can promote the peers participate in network activities share resources and restrain the "Free-Riding" behaviors.