The Research Of Trust Model And Incentive Mechanism In Mobile Peer-to-Peer Networks

Due to the attribute of the node in mobile P2 P network: autonomy and anonymity, also the topology structure of mobile P2 P network changes with the movement of nodes, the mobile network is faced with serious security problems, such as the spread of malicious files and the abuse of internet resources etc. At the same time, the nodes in MP2 P is selfish, the phenomenon of nodes’ cooperation is normal. So it is necessary to build an effective trust model and incentive mechanism to solve the problems.This paper proposes a trust model based on social trust. Aiming at the condition of the lose of trust information, the trust mode take social trust as supplement. The main advantage of the social trust is that the node can evaluate another node without the trade experience. When the trust information can be achieved, the trust can be calculated by the formula: α*directive trust value+β*-recommendation trust value. The paper calculates recommendation trust value with the number of trade and the number of successful trade. At the same time, the paper describes the relationship between the nodes with a vector.The paper proposes an incentive mechanism based on fuzzy Q-learning theory does the following work: According the different status of the nodes,the paper takes the different factors of the nodes as the input variables;initialize the fuzzy rule base; design the instant return function;describes the specific process of evolutionary game based on fuzzy Q learning. The simulation proves that the mechanism has a high convergence speed of the Q value and can effectively motivate nodes actively participate in the network activities.