Study On Global Trust Model With Recommendation In P2P Network

Peer-to-Peer systems have being experiencing a rapid growth in the past several years. In addition, its goal is that, to create an inter-connected network environment for terminal users at the edge of network, in order to meet the direct-communication among users. The open and anonymous nature of P2P networks exposes a number of security vulnerabilities, so peers are not responsible for their behaviors, and what's more, lots of malicious nodes provide faked files and malicious behaviors. Establishing a trust model in P2P system can successfully mitigate this risk by finding reliable trustworthy communication partners. Some researches show that trust-based P2P networks have great similarity with interpersonal networks. So recurring to the trust model of society network, P2P trust model can be built as society trust network.Firstly, the emergence and development of Peer-to-Peer technology are surveyed. The concept of trust and trust model of P2P technology are explained and introduced. Then currently popular P2P trust models and some relevant researches are presented and analyzed. Their shortcoming and strongpoint are brought out, and some methods to improve them are proposed.Secondly, a recommended-based global trust model is introduced, through the classic trust model and Fuzzy set theory, then give the definition of trust value and Mathematics description, verifies its rationality of our model, then we show the method of accelerating the trust value.After that, a distributed global trust value maintain strategy is introduced to achieve the global trust algorithm based on nearness-Weighted recommend- ation. The related data struct and the trust model's work flow and the trust algorithm are introduced.Lastly, the software package Query Cycle Simulator is used to achieve our experiment environment, which is developed by Stanford University P2P research group; our model is analyzed through its rationality and ability to defend malicious peer's attack.