传感技术学报2016，Vol.29Issue(10)：1579-1588,10.DOI:10.3969/j.issn.1004-1699.2016.10.019
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
摘要
Abstract
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 the structure of WSN approximately using RSSI similarity among nodes and achieve high localization accuracy ,but it only 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 using a 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 results show 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Key words
wireless sensor networks/localization/canonical correlation analysis/locality preserving projections分类
信息技术与安全科学引用本文复制引用
朱素文,曾宪华,胡梦..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[J].传感技术学报,2016,29(10):1579-1588,10.基金项目
国家自然科学（61075019）；2015年重庆市研究生科研创新项目 （）
