【作者】 赵晓东；

【导师】 熊翱；

【作者基本信息】 北京邮电大学 ， 计算机科学与技术， 2013， 硕士

【摘要】 无线传感器网络是由大量低成本传感器节点,以多跳自组织形式构成的无线网络,用来监测目标区域内的敏感信息。同传统有线网络相比,无线传感器网络工作环境恶劣、无人值守且能量有限,使得其节点发生故障的概率要比传统网络设备发生故障的概率高很多。因此,研究无线传感器网络故障检测问题,具有非常重要的意义。无线传感器网络故障检测主要指实时的监测传感器节点的工作状态,获取故障节点的信息。本文的主要研究内容和成果如下：1)针对无线传感器网络故障检测中能耗较大的问题,综合考虑传感器节点的物理构成及能耗分布情况,提出了基于邻居节点比较故障检测法的邻居节点选取算法。本文所提出的邻居节点选取算法从节省能量的角度出发,在保证一定的故障检测性能的前提下只选取当前节点周边的一些关键节点作为邻居节点,参与后续故障检测的邻居比较过程。这样既保证了故障检测的准确率,同时也避免了因故障检测消耗过多能量而缩短网络寿命的问题。2)为了充分利用传感器节点的历史经验数据,本研究借助Dirichlet分布函数,为传感器节点建立了信誉函数,提出了基于节点信誉的无线传感器网路故障检测算法。同时,算法还引入了邻居节点选取算法,只选取一些关键节点参与邻居比较的过程,大大节约了能量。故障检测算法包括检测模块、信誉更新模块和故障判定模块三部分。检测模块计算节点当前时刻的健康等级；信誉更新模块负责维护、更新节点的信誉值；故障判定模块则对是否发生节点故障和敏感事件进行判定.更多还原

【Abstract】 Wireless sensor networks often consist of a large number of low-cost sensor nodes, which are deployed in extreme environment to collect data and monitor some interest things. Due to limitations in available resources, such as power, memory and computing capability, sensor nodes, operating in an unattended mode, have higher fault rates than traditional network nodes. Therefore, the fault detection for wireless sensor networks has a very important theoretical and practical significance.Fault detection in wireless sensor network means to monitor the working condition of the sensor nodes and get some information of the failed nodes. The research content and results are as follows.1) With the limited energy in wireless sensor network, considering the physical makeup of the sensor nodes and distribution of energy consumption, we propose the neighbor nodes selection algorithm. The algorithm, starting from the point of view of energy savings, under the premise of a tolerable fault detection performance, only select some key nodes as the neighbor nodes to participate the comparison in the fault detection algorithm. The neighbor nodes selection algorithm not only ensures the accuracy of fault detection, but also avoids the excessive consumption of energy, which extend the life of the network effectively.2) In order to make full use of the historical data, the paper, based on the Dirichlet distribution function, establishes a reputation function for each sensor node and proposes the fault detection algorithm for wireless sensor networks. Besides, the fault detection algorithm only selects some key nodes to participate the comparison, which help to save energy greatly. The algorithm consists of measurement module, reputation updating module and fault judgement module. The health level of sensor node was calculated in measurement module. The reputation updating module is responsible for maintaining and updating the reputation of the sensor node. Fault judgment module determines whether a node is faulty or not, and whether the sensitive event happens or not.更多还原