基于自适应加权算法的WLAN/MARG/GPS组合定位系统  

WLAN /MARG /GPS integrated positioning system based on a self-adaptive weighted algorithm

作　　者：刘兴川[1,2] 吴振锋[2] 林孝康[1] 

机构地区：[1]清华大学电子工程系,北京100086 [2]中国电子科技集团公司第二十八研究所,南京210007

出　　处：《清华大学学报（自然科学版）》2013年第7期955-960,共6页Journal of Tsinghua University(Science and Technology)

基　　金：中国电子科技集团公司第二十八研究所博士后课题(130009)

摘　　要：随着智能手机的普及和移动互联网的成熟,基于移动位置的服务亟需一种基于智能终端、高精度、可实现室内室外无缝定位的系统。但是现有定位技术都存在各自不足,为了解决这个问题,该文设计了基于粒子滤波和Kalman滤波的自适应加权融合算法,融合了异构但互补的3种定位技术,基于智能手机实现了低成本、高精度的WLAN/MARG/GPS组合定位系统。室内外环境下的实验结果表明:该系统与WLAN/MARG和GPS/MARG定位系统相比,平均定位误差分别减小了52%和63%,并且实现了室内外的无缝定位。With the maturity of wireless-LAN （WLAN） and the popularity of intelligent terminals, accurate indoor/outdoor seamless positioning systems using smartphones are critical in mobile location based services （MLBS）. However, existing positioning systems have some disadvantages. This paper presents a self-adaptive weighted algorithm based on a particle filter and a Kalman filter that merges three heterogeneous but complementary positioning technologies. The low-cost, high-precision GPS/WLAN/MARG （magnetic, angular rate, and gravity sensors） integrated positioning system is realized in a smartphone. Tests in indoor/outdoor environments indicate that the positioning average error of this system is reduced by as much as 52% and 63% compared with WLAN/MARG and GPS/MARG positioning systems. Thus, the system provides much better seamless indoor/outdoor positioning.

关 键 词：无线局域网 组合定位 MARG传感器 粒子滤波 

分 类 号：TN929.5[电子电信—通信与信息系统]