【作者】 李建平；

【导师】 王晓凯；

【作者基本信息】 山西大学 ， 电子与通信工程， 2017， 硕士

【摘要】 随着无线传感器网络(Wireless Sensor Network,WSN)技术的快速发展,及其在军事、智能家居、电网、环境监测等领域的广泛应用,可靠性评估已成为不容忽视的关键环节之一。对WSN可靠性进行评估具有一定的理论意义和实用价值。本文分别基于可能性理论、渗流理论和模糊神经网络,建立了WSN可靠性的描述,在此基础上,构建了可靠性的评估模型,并进行了仿真实验和分析。主要内容如下:(1)基于可能性理论,将WSN可靠性定义为由传感器节点构成的状态函数不低于某一值的可能性。将传感节点寿命用可能性分布描述,通过构造函数建立传感节点寿命与WSN寿命的关系,最终构造出用于评价WSN可靠性的可靠性函数。(2)基于渗流理论,将WSN失效过程视为一个相变过程,通过相变临界阈值建立节点可靠性与WSN可靠性间的关系,得出WSN可靠性函数。(3)基于模糊神经网络,将WSN的14个重要指标作为模型的输入、可靠矩阵作为模型的输出,构建了从WSN系统层次全面评估其可靠性的神经模型。更多还原

【Abstract】 With the rapid development of wireless sensor network technology and its extensive application in military,intelligent home,power grid and environmental monitoring,WSN reliabilityevaluation hasbecome one of the key points that can not be ignored.It is of theoretical significance and practical value to evaluate the reliability of WSN.Based on possibility theory,percolation theory and fuzzy neural network,the reliability of WSN is described in this paper.On this basis,the evaluation model of reliability is constructed,and the simulation experiment and analysis are carried out.The main contents are as follows:(1)Based on possibility theory,the reliability of WSN is defined as the probability that the state function of sensor nodes is not less than a certain value.The lifetime of sensor nodes is described by probability distribution,and the relationship between the lifetime of sensor nodes and the life of WSN is constructed by the constructor.Finally,the reliability function for evaluating the reliability of WSN is constructed.(2)Based on the percolation theory,the WSN failure process is regarded as a phase transition process.The relationship between node reliability and WSN reliability is established by the critical threshold of phase transition,and the WSN reliability function is obtained.(3)Based on the fuzzy neural network,the 14 important indexes of WSN are taken as the input of the model and the reliable matrix as the output of the model.A neural model was designed to evaluate its reliability from the WSN system level.更多还原