基于子信道信息融合的室内定位方法的研究  

Study on indoor positioning method based on subchannel information fusion

作　　者：刘安 李金宝[1,2,3] 黎玲利[1,2,3] Liu An;Li Jinbao;Li Lingli(College of Computer Science and Technology, Heilongjiang University, Harbin 150080, China;Key Laboratory of Database and Parallel Computing of Heilongjiang Province Heilongjiang University, Harbin 150080, China;College of Software, Heilongjiang University, Harbin 150080, China)

机构地区：[1]黑龙江大学计算机科学与技术学院,哈尔滨150080 [2]黑龙江大学黑龙江省数据库与并行计算重点实验室,哈尔滨150080 [3]黑龙江大学软件学院,哈尔滨150080

出　　处：《仪器仪表学报》2018年第3期242-249,共8页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61370222,61602159)项目资助

摘　　要：随着基于位置的服务不断发展,室内定位逐渐成为研究的关键问题,Wi Fi设备的不断普及也为室内定位的研究提供了新的思路。由于室内环境的复杂性以及信号强度信息的不稳定性,传统的定位技术存在着较大的实验误差。利用奇异值分解（SVD）方法对子信道信息进行融合,将原始的信道信息转变为子信道信息的非线性叠加,将能够显著描述指纹的主要信息成分作为指纹特征,并对指纹位置的几何特性进行分析,利用k最近邻（k NN）算法进行指纹选取。理论分析与实验结果都证明了方法的有效性。As the location-based services are developing continuously,indoor positioning has gradually become the key issue in the research.The popularizing of Wi Fi devices also provides a new idea for indoor positioning research.Because of the complexity of the indoor environment and the unstability of the signal strength information,traditional positioning technology has large experiment error.This paper uses the SVD method to fuse subchannel state information,the original CSI value is transformed into the nonlinear superposition of a series of subchannel information.The main information components that can best describe the fingerprint are taken as the fingerprint features.The geometric characteristic of fingerprint location is analyzed,and the k NN algorithm is used to conduct the fingerprint selection and reduce the error.Both theoretical analysis and experiment result prove the effectiveness of the proposed method.

关 键 词：室内定位 信道状态信息 子信道信息融合 接入端 指纹识别 

分 类 号：TN912[电子电信—通信与信息系统] TH79[电子电信—信息与通信工程]