面向链状传感器网络的节点部署和网络路由方案研究

发布时间：2018-06-22 21:08

本文选题：传感器网络 + 负载平衡　；参考：《电子科技大学》2017年硕士论文

【摘要】：传感器网络多跳中继通信的工作方式容易导致通信负荷集中在汇聚节点附近的传感器节点中,导致这些节点能量消耗更快,形成通信热点,这个问题称为通信负载不平衡问题。窄长链状传感器网络是一种传感器节点部署呈窄长链状形式的的传感器网络,由于其特殊的结构,使其相对于常规的传感器网络更容易导致数据转发集中在部分节点造成通信负载不均衡,也更容易由部分节点耗尽能量引发网络中断,使得兼顾能量效率的负载平衡问题成为困扰窄长链状传感器网络应用的核心难题。为了解决上述难题,本文以节点部署策略和配套路由算法为核心解决这个问题:多个移动汇聚节点作为网络中间层被引入,按照规划静止和行驶。在窄长链状部署的传感器节点和移动汇聚节点的基础上,动态分簇路由算法被提出,以实现传感器节点之间和移动汇聚节点之间的通信负荷平衡,负载均衡并不以牺牲通信能量效率为代价,通信能量效率依然保持在较高的水平。传感器网络的分簇常常也会带来客观的成本,它包括簇首节点广播自身位置,簇成员接收信息,判断归属再反馈信息给簇首节点。在本文提出的方案中,所有传感器节点根据存储在自身内部的算法自动组网,无需分簇信息的广播,仅有极少的时间同步信息即可保证动态分簇的实现。相对于动态分簇,时间同步每隔很长一段时间才需要,这大大降低了建簇能量成本,同时还提高了建簇速度。为了验证该方案的有效性,本文从入侵监测的环形栅栏覆盖传感器网络、用于管道监测的线型链状传感器网络这两种典型的窄长链状传感器入手分析方案的效果,网络能耗仿真结果证明,该方案实现了传感器网络的通信能量高效和负载均衡。除此之外,以确定性部署为基础实现了低成本高质量节点定位;通过对节点进行参量定义,基站可以从接收信息中判断参与通信的传感器节点及其能耗,使得方案可以进一步应用于事件驱动和询问驱动的传感器网络;网络还可以动态调整分簇比例等方法,提高自身抗毁性和对环境的适应性。
[Abstract]:The mode of multi-hop relay communication in sensor networks easily leads to the communication load concentrated in the sensor nodes near the convergent nodes, which leads to the faster energy consumption of these nodes and the formation of communication hot spots. This problem is called the problem of communication load imbalance. Narrow long chain sensor network is a kind of sensor network in which sensor nodes are deployed in narrow long chain form. Because of its special structure, Compared with conventional sensor networks, it is easier to cause data forwarding to focus on some nodes and cause communication load imbalance, and it is also easier to cause network disruption by some nodes running out of energy. Therefore, load balancing with energy efficiency is the core problem for narrow long chain sensor networks. In order to solve the above problems, this paper focuses on the node deployment strategy and the matching routing algorithm: multiple mobile convergence nodes are introduced as the intermediate layer of the network, and they are stationary and driving according to the plan. Based on the narrow long chain deployed sensor nodes and mobile convergence nodes, a dynamic clustering routing algorithm is proposed to balance the communication load between sensor nodes and mobile convergence nodes. Load balancing is not at the expense of communication energy efficiency, which remains at a higher level. The clustering of sensor networks often brings the objective cost, which includes broadcasting the location of the cluster head node, receiving the information from the cluster member, judging the home and feedback information to the cluster head node. In the scheme proposed in this paper, all sensor nodes automatically network according to the algorithm stored in their own, without the broadcast of clustering information, only a little time synchronization information can ensure the realization of dynamic clustering. Compared with dynamic clustering, time synchronization is needed every long time, which greatly reduces the energy cost and improves the speed of cluster building. In order to verify the effectiveness of the proposed scheme, this paper analyzes the effects of the two typical narrow and long chain sensor networks, namely, the ring fence covering sensor network for intrusion detection and the linear chain sensor network for pipeline monitoring. The simulation results of network energy consumption show that the proposed scheme achieves communication energy efficiency and load balancing of sensor networks. In addition, on the basis of deterministic deployment, the location of nodes with low cost and high quality is realized. By defining the parameters of nodes, the base station can judge the sensor nodes involved in the communication and their energy consumption from the received information. The scheme can be further applied to event-driven and interrogation-driven sensor networks, and the network can dynamically adjust the clustering ratio to improve its own invulnerability and adaptability to the environment.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【参考文献】