一种面向不规则的无线电传播模型下的节点定位算法  

A Node Localization Algorithm for Irregular Radio Propagation Pattern

作　　者：葛文庚[1] 陈萍[1] GE Wengeng;CHEN Ping(College of Information Engineering,Huanghuai University,Henan Zhumadian 463000,China)

机构地区：[1]黄淮学院信息工程学院,河南驻马店463000

出　　处：《弹箭与制导学报》2021年第2期15-19,共5页Journal of Projectiles,Rockets,Missiles and Guidance

基　　金：河南省科技攻关项目(212102210515,182102210411,182102310783);2018年度河南省重大科技专项(182102210100)资助。

摘　　要：传统定位算法假定节点的RPP是圆形的,但由于信号衰减、阴影以及干扰等原因,节点不具有圆形传播特性。为此,提出面向非圆形RPP的基于人工神经网络ANNs的加权质心定位算法ANNs-WCL。ANNs-WCL定位算法通过新的测距算法估计未知节点与锚节点间的距离,再利用ANNs修正非圆形RPP所产生的测距误差,并由加权质心定位算法估计未知节点位置。仿真结果表明,与传统的非测距定位算法相比,ANNs-WCL算法的根均方定位误差得到有效的下降。Due to real-world phenomena such as fading, shadowing and interference, nodes in wireless sensor networks have an irregular radio propagation pattern(RPP). However, traditional localization algorithms rely on the assumption that nodes have a circular RPP. Therefore, artificial neural networks-based weighted centroid localization(ANNs-WCL) in the presence of irregular radio propagation pattern is proposed in this paper. In ANNs-WCL algorithm, it develops a new distance estimation approach able to efficiently derive distances’ estimates in closed form. Then, exploiting ANNs, we also develop distance estimation correction. Finally, the unknown node coordinates is calculated by weighted centroid localization. Simulation results show that of ANNs-WCL algorithm outperform than centroid localization algorithm in root mean square error.

关 键 词：无线传感网络 非测距定位 无线电传播模型 人工神经网络 质心定位算法 

分 类 号：TN011[电子电信—物理电子学] TP183[自动化与计算机技术—控制理论与控制工程]