文章基本信息

标题：A NOVEL HYBRID LOCALIZATION METHOD FOR WIRELESS SENSOR NETWORK
本地全文：下载
作者：Wang Jun ; Zhang Fu ; Ren Tiansi 等
期刊名称：International Journal on Smart Sensing and Intelligent Systems
印刷版ISSN：1178-5608
出版年度：2016
卷号：9
期号：3
页码：1323-1340
出版社：Massey University
摘要：Wireless sensor network is a kind of brand-new information acquisition platform, which isrealized by the introduction of self-organizing and auto-configuration mechanisms. Node localizationtechnology represents a crucial component of wireless sensor network. In this paper, a localizationmethod based on kernel principal component analysis and particle swarm optimization backpropagation algorithm is carefully discussed. First of all, taking KPCA as the front-end system toextract the main components of the localization information, and then regarding the nonlinearprincipal components extracted from distance vectors as the input samples, and meanwhile taking thecoordinates of vertices in addition to the region boundary as the output samples, the PSO-BP neuralnetwork is trained to achieve the localization model. Finally the localization of unknown nodes can beestimated. The simulation experiment result showed that the method has high ability of stability andprecision, and meets the practical need of localization.
关键词：Wireless sensor network; localization; kernel principal component analysis; particle swarm;optimization; back-propagation algorithm