无线传感器网络DV-hop定位算法分析及改进研究

发布时间：2018-04-17 04:19

本文选题：无线传感器网络 + DV-hop定位算法　；参考：《南京大学》2014年硕士论文

【摘要】：无线传感器网络(Wireless Sensor Networks, WSN)由大量廉价的微型传感器节点组成,节点之间通过无线通信方式形成多跳自组织网络。无线传感器网络在许多领域有广泛的应用,网络中节点的定位技术是无线传感器网络研究的众多关键性问题之一,本文将围绕定位问题做进一步的分析和研究。无线传感器网络中的节点分为两种：信标节点和未知节点。已知自身坐标信息的节点称为信标节点；未知节点需通过网络的连通性、节点间的距离等信息进行自身定位,获得自身坐标信息。目前已存在多种定位算法,这些算法可依据不同的标准进行分类,如可分为基于测距的定位算法和基于非测距的定位算法。在多种定位算法中,本文重点分析了DV-hop定位算法,该算法属于非测距算法中的一种典型算法。非测距算法不直接计算节点间的距离,而是依据节点间的跳数、跳距信息估算节点间的距离,再对未知节点进行自定位。算法由于无需直接测量距离信息,降低了对节点硬件的要求,节约了网络成本,但是定位精度不高。本文在对DV-hop定位算法分析的过程中,做了大量仿真实验,以平均定位误差或定位精度作为算法评价标准,分析仿真结果,并提出改进思路：在跳数信息不变的情况下,通过优化跳距信息提高算法的定位精度。两种改进算法分别为：DV-hop-NMean定位算法和DV-hop-NHs定位算法。DV-hop-NMean定位算法通过取一些特定值的均值作为跳距优化信息,DV-hop-NHs定位算法则是通过对两个特定值的加权作为跳距优化。考虑到实验的可靠性,在相同仿真环境条件下,对原DV-hop定位算法和改进定位算法分别作大量实验,将多次实验定位精度的平均值作比较,证明改进算法的可行性。两种改进算法从定位精度看,均优于原算法,定位误差可达到6.6米,降低了15厘米左右。进一步对估计的未知节点坐标做优化,采用节点坐标计算方法中的极小极大定位算法思想,分别对原DV-hop定位算法和两种改进定位算法做优化,通过大量仿真实验,将优化前后定位精度值作对比,发现在通信半径为20米不变的情况下,定位精度可以提高近1米,证明了优化算法的有效性。相关研究对无线传感器网络的定位技术及应用有借鉴意义。
[Abstract]:Wireless Sensor Networks (WSNs) are composed of a large number of cheap micro sensor nodes, which form multi-hop ad hoc networks through wireless communication.Wireless sensor networks (WSN) have been widely used in many fields. Node localization is one of the key problems in wireless sensor networks (WSN).There are two kinds of nodes in wireless sensor networks: beacon nodes and unknown nodes.The nodes that know their coordinate information are called beacon nodes, and the unknown nodes need to locate themselves through the network connectivity and the distance between nodes to obtain their own coordinate information.At present, there are many localization algorithms, which can be classified according to different criteria, such as location algorithm based on ranging and location algorithm based on non-ranging.Among various localization algorithms, this paper focuses on the analysis of DV-hop localization algorithm, which is a typical non-ranging algorithm.The non-ranging algorithm does not directly calculate the distance between nodes, but estimates the distance between nodes according to the hops between nodes and the information of hops, and then self-locates the unknown nodes.Because the algorithm does not need to measure the distance information directly, it reduces the requirement of node hardware and saves the network cost, but the positioning accuracy is not high.During the analysis of DV-hop localization algorithm, a lot of simulation experiments have been done in this paper. The average positioning error or positioning accuracy is taken as the evaluation standard of the algorithm, and the simulation results are analyzed, and the improved thinking is put forward: under the condition of invariant hops information,The location accuracy of the algorithm is improved by optimizing the hopping information.The two improved algorithms are respectively the DV-hop-NMean location algorithm and the DV-hop-NHs localization algorithm .DV-hop-NMean localization algorithm. The DV-hop-NMean localization algorithm takes the mean value of some specific values as the hopping optimization information and the DV-hop-NHs localization algorithm is optimized by weighting the two specific values as the hopping distance optimization.Considering the reliability of the experiment, a large number of experiments have been done on the original DV-hop localization algorithm and the improved localization algorithm under the same simulation environment. The average value of the multiple experiments' positioning accuracy has been compared, and the feasibility of the improved algorithm has been proved.The two improved algorithms are superior to the original algorithm in terms of positioning accuracy, and the positioning error can reach 6.6 meters, which is about 15 cm lower than that of the original algorithm.Furthermore, the unknown node coordinates are optimized and the original DV-hop location algorithm and two improved localization algorithms are optimized by using the idea of minimax localization algorithm in the node coordinate calculation method, and a large number of simulation experiments are carried out.By comparing the positioning precision values before and after optimization, it is found that the positioning accuracy can be improved by nearly 1 meter when the communication radius is 20 m constant, which proves the effectiveness of the optimization algorithm.The related research has the reference significance to the wireless sensor network localization technology and the application.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5;TP212.9

【参考文献】