对等网(P2P)系统因其开放性和匿名性等特点易受到恶意攻击和非法滥用,建立基于社区的信任模型是一个行之有效的解决途径.而现有的模型忽略了节点的动态性、活跃度及影响范围.该文在分析了P2P用户模型后,提出一种基于拓扑势的P2P社区推荐信任模型,综合评估了节点的影响力、交易量及信任度.在该模型中,分别建立了社区内部和跨社区信任度计算机制,给出了超级节点评选算法.最后的仿真实验结果验证了该信任模型的有效性和鲁棒性.%The features of autonomy, anonymity and distribution make the P2P system vulnerable to malicious attack and abuse. A feasible resolution in such an open environment is to exploit a community-based trust model to build trust relationship between peers. However, the existing model ignores the dynamic feature, the scope of activity and the influence of peers. After analyzing the P2P user model, a topological potential based recommendation trust model is proposed to integrate the influences, transactions, and reputations of nodes. In this model, the trust metrics are divided into intra- and inter-community computing mechanism. Moreover, the algorithm of selecting super node is presented. Simulation results show that the proposed trust model is effective and robust.
展开▼
