A topological potential weighted community-based recommendation trust model for P2P networks

Peer-to-Peer (P2P) technology has experienced its great popularity for resource sharing in current network services. However, lacking of prior experience and knowledge about other peers' reputation, P2P suffer from malicious behaviors launched by malicious peers. Developing effective strategies to establish trust model for the P2P system is a good way to address this problem. In this paper, we propose a novel topological potential weighted community-based recommendation trust (TPCommuTrust) model architecture, which is based on topological potential weighted community credential to score and rate peers' previous data transactions. Contributed by topological potential, which characterizes repercussion effect between peers, the reputation model of the peers is created. The unique and efficient mechanism is proposed to perform data transaction based on credentials intra and inter-community. Moreover, how TPCommuTrust can provide an access control against malicious peers is illustrated as well. Simulation results show that TPCommuTrust model is robust and highly efficient to identify malicious nodes.