【作者】 冯芳；

【导师】 程良伦；

【作者基本信息】 广东工业大学 ， 控制理论与控制工程， 2011， 硕士

【摘要】 目前,无线传感器网络技术在军事侦察、环境科学、医疗卫生、工业自动化以及地质灾害等领域得到了广泛的应用。在基于无线传感器网络的农田土壤污染监控系统中,由于传感器节点数量众多,能量有限,分布区域广,而且部署区域环境复杂,所以研究无线传感器网络的节能机制具有非常重要的意义。本文针对系统的低功耗要求,重点研究无线传感器网络中的分簇路由协议,目的是减少网络中节点的能耗,延长监控区域内无线传感器网络的寿命。主要工作为以下六个方面：1.对无线传感器网络中分簇路由协议和数据融合技术进行分析,总结现有的研究成果,并对它们进行分类比较。2.对路由算法的应用对象——农田土壤监测系统的组成部分进行设计。3.针对区域内簇间通信的不足,提出一种高节能分层路由协议,该协议主要是根据能量因子来选择簇间通信的中间节点,达到减少簇头能耗的目的。4.针对分簇路由协议中簇头选择算法的不足,提出一种基于神经网络的自适应路山算法,该算法是在基站上采用神经网络的自适应学习来选择簇头,并根据最优权函数值选择最短路径中的下一跳节点,同时添加一个新因子来选择网关节点以进行簇头间的通信,减少了簇头节点导致网络出现瓶颈的概率5.考虑到前面两种协议只适合静态路由,且构建路由时需知道节点的全部信息,提出一种基于簇势场的数据融合策略,该策略在簇内引入序列势场,同时在簇头与Sink间引入混合势场以进行数据融合,簇头可根据其局部信息快速构建一条路由,因此可大大减少簇头的能耗,延长无线传感器网络的寿命。6.通过MATLAB和NS2分别对上述三种协议的网络生存期,节点能耗等情况进行仿真。仿真结果表明本文提出的三种节能路由协议提高了无线传感器网络节点的效率,可应用于农田土壤监测系统,并为无线传感器网络节能机制领域的研究提供了有益的探索。更多还原

【Abstract】 At present, Wireless Sensor Networks have been widely used in military reconnaissance, environmental science, medicine and health, industrial automation, geological disasters and so on. In farmland soil monitor control system based on Wireless Sensor Networks, the numbers of sensor nodes are numerours, in which the energy is limited, and the regional environmen is complex, so the study of the energy-efficiency mechanism in Wireless Sensor Networks has been of the vital significance. The thesis focused on the routing protocol based on clustering in Wireless Sensor Networks combinging the low power requirements of the system. The aim was to reduce the energy consumption of nodes and extend the lifetime of Wireless Sensor Networks.The main contributions included the following:Firstly, the routing protocols based on clustering and the data fusion technologies in Wireless Sensor Networks were analyzed. It summarized researching status in the current, classified and compared them.Secondly, the farmland soil monitor system was designed.Thirdly, a highly energy-efficient hierarchical routing protocol was proposed to solve the lack of communication between the cluster heads, which selected candidate nodes according to a new factor.Fourthly, for the lack of clustering algorithm, this thesis presented an adaptive routing algorithm based on neural network, which selected the cluster head by the adaptive learning of neural network on Base Station, choosed the next hop in the shortest path by the optimal value of the weighting function and added a new factor to select the gateway nodes for communication among cluster heads. So it avoided from the cluster head nodes becoming a network bottleneck.Fifthly, to solve the deficiency of being suitable for static routing and holding the all the information of nodes to rout, a data aggregation scheme based on potential field of clustering was proposed. It introduced potential field into the cluster and hybrid potential field between the sink and cluster heads (CH) to aggregate data. What’s more, CH could make full use of the local information to make the routing decisions rapidly. So it could reduce the energy consumption of cluster heads and extend the lifetime of Wireless Sensor Networks.Sixthly, the three protocols were simulated in network lifetime and energy consumption of nodes by MATLAB and NS2, respectively. The results shew that the proposed energy efficient routing protocol in further improved energy efficiency of the operation mechanism in Wireless Sensor Networks. So the thesis provided useful exploration for the study of energy mechanism in Wireless Sensor Networks.更多还原