改进的局部保持典型相关分析的无线传感器网络节点定位方法  

Improved Locality Preserving Canonical Correlation Analysis Node Localization Method in Wireless Sensor Networks

作　　者：朱素文[1,2] 曾宪华[1,2] 胡梦[1,2] ZHU Suwen ZENG Xianhua HU Meng(College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China Chongqing Key Laboratory of Computational Intelligence, Chongqing 400065, China)

机构地区：[1]重庆邮电大学计算机科学与技术学院,重庆400065 [2]计算智能重庆市重点实验室,重庆400065

出　　处：《传感技术学报》2016年第10期1579-1588,共10页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金项目(61075019);2015年重庆市研究生科研创新项目(CYS15173)

摘　　要：利用接收信号强度(RSSI)进行无线传感器网络(WSN)定位是一类低成本定位方法。局部保持典型相关分析定位(LE-LPCCA)算法能通过节点间RSSI数据的相似度信息近似拟合WSN结构,取得了较高定位精度。但该算法只使用节点间相似性信息未保留信号空间和物理空间的相关性信息,且求解未知节点坐标时使用粗糙的质心法。针对以上问题,提出改进的局部保持典型相关分析定位(LE-ILPCCA)算法,该算法在样本训练阶段用平衡参数将数据的相似性和相关性信息进行融合,求取RSSI内在低维坐标表示的投影变换;在定位阶段,求解已知节点位置坐标和RSSI内在低维坐标之间存在的线性转换关系,获得未知节点的坐标。实验结果表明,本文算法与LE-LPCCA和LE-CCA相比定位精度高、稳定性强。Location methods in wireless sensor networks（WSN）by received signal strength indicator（RSSI）are rela-tively inexpensive. Location Estimation-Locality Preserving Canonical Correlation Analysis（LE-LPCCA）can fit thestructure of WSN approximately using RSSI similarity among nodes and achieve high localization accuracy,but itonly uses data similarity while ignoring data dependency between signal and physical space,and employs rough cen-troid method. As to problems above,this paper proposes Location Estimation-Improved Locality Preserving Canoni-cal Correlation Analysis（LE-ILPCCA）method. In training phrase,it combines data similarity and dependency usinga balance parameter to compute more precise projection transformation of RSSI inner lower dimensional coordi-nates;in localization phrase,it calculates location of unknown nodes utilizing accurate transformational relation be-tween position coordinates and RSSI inner lower dimensional coordinates of known nodes. Experimental resultsshow that our method has a higher accuracy and stability than LE-LPCCA and LE-CCA.

关 键 词：无线传感器网络 定位 典型相关分析 局部保持投影 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]