削减NLOS误差的UWB室内定位算法  

UWB interior positioning algorithm for NLOS error reduction

作　　者：李新春[1] 阳士宇 张玉琛 孙鹤源 LI Xinchun;YANG Shiyu;ZHANG Yuchen;SUN Heyuan(School of Electronics and Information Engineering,Liaoning Technical University,Huludao 125105,P.R.China;Graduate School,Liaoning Technical University,Huludao 125105,P.R.China)

机构地区：[1]辽宁工程技术大学电子与信息工程学院,辽宁葫芦岛125105 [2]辽宁工程技术大学研究生院,辽宁葫芦岛125105

出　　处：《重庆邮电大学学报（自然科学版）》2022年第5期758-765,共8页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：国家自然科学基金(61372058)。

摘　　要：为了解决非视距(non-line of sight,NLOS)环境下超宽带(ultra-wideband,UWB)室内定位精度低的问题,提出一种基于到达时间差(time difference of arrival,TDOA)的UWB室内定位算法来削减NLOS误差,提高定位精度。利用卡尔曼滤波算法,初次估计移动标签节点与各锚节点间的TDOA值;基于TDOA差值进行NLOS误差判别,依据判别结果对受NLOS误差影响的测量值进行视距(line of sight,LOS)重构,并利用仅受LOS误差影响的历史值更新卡尔曼滤波的协方差阵;利用Chan-Taylor算法进行迭代计算,得到最终精确的定位结果。实验结果表明,在LOS环境下,提出的算法能达到分米级的定位精度;在受NLOS影响的环境下,该算法能有效提高定位精度,减小定位误差,具有更好的稳定性。A TDOA-based ultra-wideband(UWB)interior positioning algorithm is proposed to reduce the non-line of sight(NLOS)error and enhance the positioning accuracy in order to resolve the problem of poor UWB indoor positioning accuracy in the NLOS environment.In this algorithm,the time difference of arrival(TDOA)values between mobile labeled nodes and anchor nodes are initially estimated using the Kalman Filter algorithm,and then the NLOS error is determined based on the TDOA values.The line of sight(LOS)measurements affected by the NLOS error are reconstructed based on the result determined,and the covariance matrix in the Kalman Filter is updated using the historical values only affected by the LOS error.Lastly,iterative computations are conducted with the Chan-Taylor algorithm to obtain the final accurate positioning result.The experimental results show that this algorithm can achieve decimeter level positioning accuracy in LOS environment and can effectively help enhance the positioning accuracy in the NLOS environment,reducing the positioning error and demonstrating better stability.

关 键 词：非视距 超宽带 室内定位 到达时间差 卡尔曼滤波 

分 类 号：TN92[电子电信—通信与信息系统]