A trust model based on fuzzy Q-learning algorithm in mobile P2P network

The main function of MP2P network is promoting the collaboration between mobile devices and the resources sharing by direct data exchange. In this paper, we propose a trust model based on fuzzy Q-learning algorithm (FQTM). The FQTM model can maximize mobile nodes´ benefits by learning and adapting to unfamiliar environments. This model applies Fuzzy Theory and Game Theory to the calculation process of agents´ action-state transition probability. Mobile nodes adjust their decisions constantly based on past records with the policy iteration algorithm, then FQTM updates Q value until it converges to a relatively stable statement. Experiments show that FQTM has a faster speed than greedy strategy of Q-value convergence and help mobile agents to maximize their benefits.