Abstract: To attack the problem of target tracking in LOS/NLOS hybrid environments, a robust Interactive Multiple model（IMM） approach based on fused Time of Arrival（TOA） and Received Signal Strength Indicator（RSSI） measurements is proposed in this paper. Extended Kalman Filter（EKF） and Extended H-infinity Filter（EHF） is respectively used to describe the LOS and NLOS transmission, which is modeled by a Markov process. Monte Carlo simulation results show the practice and effectiveness of the proposed algorithm. It has higher positioning accuracy and better tracking stability with similar computing complex compared with the TOA based approach.

Key words: target tracking, extended H-infinity filter, extended Kalman filter, interacting multiple model, LOS/NLOS hybrid environment, TOA-RSSI

摘要： 针对视距（Line-of-sight，LOS）和非视距（None-line-of-sight，NLOS）混合环境中的运动目标跟踪问题，提出一种基于TOA（到达时间）与RSSI（接收信号强度）测量融合的交互式多模型（Interacting Multiple Model，IMM）鲁棒跟踪算法。目标与基站之间的LOS、NLOS传输分别用扩展卡尔曼滤波（EKF）和扩展[H∞]滤波（EHF）进行匹配，并采用马尔可夫过程对模型间的转换进行描述。Monte Carlo仿真结果表明，与单纯TOA测量跟踪相比，该算法具有较高的定位精度和较好的跟踪稳定性，且计算复杂度相当，具有较好的可实现性。

关键词: 目标跟踪, 扩展[H&infin, ]滤波, 扩展卡尔曼滤波, 交互式多模型, LOS/NLOS混合环境, TOA-RSSI