基于Wright-Fisher过程的WSNs节点信任随机演化策略

摘要

为解决无线传感器网络节点间的信任影响节点协作的问题,考虑到节点数量有限及个体随机性,基于Wright-Fisher过程的随机演化博弈,提出WSNs节点信任随机演化策略,并加入与节点信任相关的惩罚机制.该策略弥补了复制子动态不适用于节点数量有限的WSNs节点信任演化建模问题,经过随机动力学分析,推导并证明了达到演化稳定状态的定理.最后通过实验验证定理并分析惩罚力度和选择强度对演化稳定状态的影响.%To solve the problem of trust decisions among WSNs nodes which affects the cooperation between nodes, considering limited numbers of nodes and individual stochastic effect, a stochastic evolutionary trust strategy of WSNs nodes based on Wright-Fisher process is proposed, and a punishment mechanism related to the trust level of nodes is also introduced.The strategy remedied the defect that replicator dynamics was inapplicable to the WSNs due to the limited numbers of nodes.Theorems of arriving at the evolutionary stable state were deduced and proved through the analysis of stochastic dynamics.Experiments verified the conclusions and analyzed the impacts of punishment levels and selection intensities.