无线传感器网络中蒙特卡洛定位算法的研究  

Study on Monte Carlo Location Algorithm in Wireless Sensor Networks

作　　者：张绮曼 张颖[1] ZHANG Qi-man;ZHANG Ying(College of Information Engineering,Shanghai Maritime University,Shanghai 201306,China)

机构地区：[1]上海海事大学信息工程学院,上海201306

出　　处：《计算机科学》2018年第12期77-80,116,共5页Computer Science

基　　金：国家自然科学基金项目(61673259);上海市科委国际学术合作交流项目(15220721800)资助

摘　　要：在无线传感器网络的节点定位领域,常用的以蒙特卡洛为基础的定位算法均存在定位误差大、采样效率低的问题。为了提高无线传感器网络中针对移动节点的采样效率和定位精确度,文中采用马尔科夫链进行抽样,提出了一种基于蒙特卡洛的改进算法。该算法在蒙特卡洛算法的基础上,结合马尔科夫链采集节点样本,随后对其进行过滤,再通过对得到的节点位置值进行加权计算,得到节点的准确位置。仿真实验结果表明,通过该算法得到的节点定位误差低于其他算法,提高了采样效率以及对移动节点的定位准确率。In the field of node location of wireless sensor network,there exist problems of high location error and low sampling efficiency of the commonly used Monte Carlo-based location algorithm.In order to improve the sampling efficiency and location accuracy of mobile node in wireless sensor networks,this paper made use of Markov chain to sample,and proposed an improved location algorithm based on Monte Carlo.The new algorithm combines the Markov chain to complete the collection of node samples based on Monte Carlo algorithm,then filters them,and finally obtains the exact position of the node by weighting the obtained node position values.Simulation results indicate that the proposed algorithm has lower location error than other algorithms,and improves the sampling efficiency and location accuracy for moving nodes.

关 键 词：无线传感器网络 蒙特卡洛算法 马尔科夫链 移动节点定位 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]