基于IMM-KF的混合场景非视距定位算法  

Hybrid non-line-of-sight localization algorithm based on IMM-KF

作　　者：黄庆东 张典 李佳欣 郭振 王皓 HUANG Qingdong;ZHANG Dian;LI Jiaxin;GUO Zhen;WANG Hao(School of Communications and Information Engineering,Xi’an University of Posts and Telecommunications,Xi’an 710121,China;Shaanxi Key Laboratory of Information Communication Network and Security,Xi’an 710121,China)

机构地区：[1]西安邮电大学通信与信息工程学院,陕西西安710121 [2]陕西省信息通信网络及安全重点实验室,陕西西安710121

出　　处：《西安邮电大学学报》2024年第3期12-19,共8页Journal of Xi’an University of Posts and Telecommunications

基　　金：国家自然科学基金项目(62101442)。

摘　　要：为了降低视距(Line of Sight,LOS)和非视距(Non-Line of Sight,NLOS)混合场景下无线定位的误差,提出一种基于交互式多模型-卡尔曼滤波(Interactive Multiple Model-Kalman Filter,IMM-KF)的残差选择NLOS定位算法。构建适用于LOS的多边定位模型和适用于NLOS的残差选择3边定位模型,通过似然概率加权估计两个模型的融合位置。结合卡尔曼滤波进行误差估计确定残差,得到最优的位置估计值,从而降低计算复杂度和由计算不准确导致的模型失配问题。仿真结果表明,所提算法在混合场景中NLOS噪声服从高斯分布、指数分布及均匀分布下,定位精度优于其他对比算法,能有效降低NLOS误差。To reduce the error in wireless positioning in mixed line of sight(LOS)and non-line of sight(NLOS)scenarios,a residual selection NLOS positioning algorithm based on the interactive multiple model-Kalman filter(IMM-KF)is proposed.This approach constructs a multi-lateration positioning model suitable for LOS and a residual selection trilateration positioning model suitable for NLOS.The fusion position of these two models is estimated through likelihood probability weighting,and the residuals are determined by the Kalman filtering for error estimation.The optimal position estimate is obtained,reducing both computational complexity and model mismatch issues caused by inaccurate calculations.Simulation results demonstrate that the proposed algorithm effectively reduces the NLOS errors under noises with Gaussian distribution,exponential distribution,and uniform distribution,and achieves higher positioning accuracy compared to the other algorithm.

关 键 词：无线传感器网络 非视距 视距 交互式多模型 卡尔曼滤波 残差选择 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]