A Layer Trust Model Based On Chord In P2P Network

Peer-to-Peer (P2P) has become one of the most important applications in internet. However, the nature of openness, anonymity and loose-coupling relationship between peers in P2P networks leads to a complete lack of responsibility for the content a peer puts on the network, opening the door to abuses of these networks by malicious peers. Moreover high dynamics of Peer behaviors always bring on unsteadiness of the network bandwidth and Performance, which restricted the P2P further development seriously. Therefore, to improve the Performance of P2P networks, the peers' voluntary operations must be taken into account as well as the incentive mechanisms which incent peers cooperate and consume resources rationally. This thesis studies the problems of unrealizable services and uncooperation in P2P networks, and results in the following achievements.Based on analysis and research works on current trust mechanisms and Market mechanisms, this thesis propose a Layer Trust Model based on Chord in P2P Network. Nodes in this model are divided into two parts and are dealt with separately by node's heterogeneity (processing ability and behavior style). Meanwhile this thesis design two mechanisms: bi-directional mechanisms based on trust and nodes evaluation mechanism like as profit sharing in market mechanism. It could access the information efficiently and alleviate flow impact which brought by node's dynamic,autonomy,and flow moment widely exist in trust model base on Recommendation and prevent the malicious peers and incentive peer provide better services, thus boost the performance of network.This thesis simulates and evaluates TMCL and other trust model from the network traffic, the peers' betweenness centrality, the successful request rate and the successful of download rate through the Query Cycle simulator. The Simulation results show that the model of TMCL has higher success rate of transaction, and better performance in avoiding network resource waste, reducing network flow and incentiving peers.