基于KPCA的无线层析成像定位方法  

Wireless Tomography Positioning Method Based on KPCA

作　　者：陆俊[1] 柯炜[1,2] 金杰 陈梦玲[1] 王彦力 左浩然[1] Lu Jun;Ke Wei;Jin Jie;Chen Mengling;Wang Yanli;Zuo Haoran(Jiangsu Key Laboratory on Opto-electronic Technology,School of Physics Science and Technology,Nanjing Normal University,Nanjing 210023,China;Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application,Nanjing 210023,China)

机构地区：[1]南京师范大学江苏省光电子技术重点实验室,物理科学与技术学院,江苏南京210023 [2]江苏省地理信息资源开发与应用协同创新中心,江苏南京210023

出　　处：《南京师大学报（自然科学版）》2020年第1期31-39,共9页Journal of Nanjing Normal University(Natural Science Edition)

基　　金：2018年江苏省研究生科研创新项目(KYCX18_1187).

摘　　要：无线层析成像(radio tomographic imaging,RTI)技术作为无设备目标定位(device-free localization,DFL)的主要方式之一,在被定位目标不携带任何定位装置的情况下仍能实现定位,具有广泛的应用前景.但由于接收信号强度(received signal strength,RSS)信息容易受到环境变化和噪声的影响,RTI成像图上往往不可避免地存在着背景噪点,有时甚至还有伪目标出现在图像上.为了提高RTI成像质量,本文提出一种基于核主成分分析(kernel principal component analysis,KPCA)的增强型RTI方法,该方法利用KPCA的学习能力来提取有效受目标影响的链路特征信息,从而达到克服噪声影响和提高定位精度的目的.室内外实验结果表明,该方法的成像质量和定位精度都要优于现有RTI方法.Radio Tomography Imaging(RTI)technology is one of the main methods of device-free localization(DFL).It can still achieve positioning without being positioned with any positioning device,and with a wide range of application prospects.However,since the Received Signal Strength(RSS)information is easily affected by environmental changes and noise,background noise is inevitably present on the RTI image,and sometimes even pseudo targets appear on the image.In order to improve the quality of RTI,this paper proposes an enhanced RTI method based on Kernel Principal Component Analysis(KPCA),which uses the learning ability of KPCA to extract the link characteristic information that is effectively affected by the target,in order to overcome the effects of noise and improve positioning accuracy.The results of indoor and outdoor experiments show that the imaging quality and positioning accuracy of this method are better than the existing RTI method.

关 键 词：无线层析成象 无设备定位 接收信号强度 核主成分分析 

分 类 号：O451[理学—无线电物理]