The Research On Trust Model For P2P Network

The equal nodes in P2P network are both client and server, what making the P2P network technology apply in a widely range. But the characteristics of opening, anonymity, autonomy, and the lack of effective constraints on the nodes behavior in P2P network, what cause the lacking of trust between nodes, and even badly restrain the further development of P2P network technology. Now more effective approach is to build trust model to manage the trust between nodes in the P2P network, and many studies are directed at the local reputation of distributed model, because it has a strong flexibility performance in the calculation of trust between nodes.Considering the shortcomings of current trust models, this paper presents P2P Trust Management model based on a recommendation evidence and risk assessment RTM. When calculating the trust of nodes, this paper consider the local reputation, the recommendation reputation, and introduce the risk factors into the reputation calculation. In this paper, we use the decay function to describe the decay nature of trust value about time much more exactly, and define those functions about local reputation risk, recommendation reputation risk of the nodes which integrate with the risk factors. Using the improved D-S rules to together the trust value of recommend, which effectively solve the conflict problem in recommendation; and considering the security of the model, this paper utilize the law of deviation from center effectively eliminate some of recommended nodes which are inefficient, abnormal, and effectively resist the attacks of malicious nodes.In the simulation, in order to better analyze the performance of the model, this paper realize the part of PeerTrust and PET mechanism, and the RTM model relatively do well in containing the malicious nodes'attacks, and show advantage in the simulation experiments such as simple attacks slander attac,simple malicious gang attack.