一種強健擴展型卡爾曼濾波為基礎的交互多模型算法用於改善在隨機非直視環境下定位性能

本論文提出一種強健擴展卡爾曼濾波器為基礎的交互多模型算法(IMM-REKF)，用於改善在隨機非直視距(NLOS)傳播環境中，非直視距量測訊號所帶來的嚴重定位誤差。為了可以更切確的模擬出真實的環境，論文中我們以不同的NLOS發生機率去做模擬，且假設為隨機分佈的情況。我們以直視距(LOS)基地台的資訊當作參考的基準提出量測限制的架構，濾掉不合理的量測訊號。為了探討智能性對強化IMM-REKF性能的可能性，本論文亦根據環境的變動加入模糊推論系統去調整過程干擾(process noise)協方差矩陣，提出模糊IMM-REKF算法(FIMM-REKF)。在最後的模擬結果中，顯示出我們所提出的方法是可以有效的改善定位的誤差，成功的抑制NLOS訊號所帶來的影響，尤其是在高NLOS發生機率的環境底下，還是能夠符合我們所預期的規範，具有很好的穩定性和強健性，遠優於文獻中R-IMM的算法表現。而論文中所提出的 FIMM-REKF算法表現其實和IMM-REKF算法差不多，但前者運算所耗費的時間明顯的高出許多，權衡之下還是以IMM-REKF的算法為主。

關鍵字

交互多模型算法；到達時間法；擴展型卡爾曼濾波器； M-estimation ；移動定位；模糊推論

並列摘要

In this thesis, a robust EKF-based interacting multiple-model (IMM-REKF) algorithm is proposed for significantly reducing positioning errors which are caused by NLOS measurement signal propagations in random NLOS environments. Specifically, we consider positioning environments with different NLOS occurrence probability and assume probability NLOS distributions are unknown. We propose a validated measurement structure based on information from LOS base stations to cut off unreasonable received measurement signals. Furthermore, we employ fuzzy inference system for tuning process noise covariance in IMM-REKF to yield a FIMM-REKF algorithm. In simulations, our proposed methods remarkably outperform the R-IMM in the literature. They reduce positioning errors by mitigating NLOS effects. Even for environments with NLOS occurrence probability, the performance of our proposed methods can still meet the FCC requirements. The IMM-REKF and the FIMM-REKF perform similarly. The IMM-REKF is a better choice because the FIMM-REKF requires more computation.

並列關鍵字

interacting multiple-model algorithm ； time of arrival ； extended Kalman filter ； M-estimation ； mobile positioning ； fuzzy inference

參考文獻

[1] S. S. Wang, M. Green, and M. Malkawa, “E-911 location standards and location commercial services,” in Proc. IEEE Emerging Technologies Symp.: Broadband, Wireless Internet Access, 2000

[3] J. Goslinski, M. Nowicki, and P. Skrzypczynski, “Performance comparison of EKF-based algorithms for orientation estimation on Android platform,” IEEE Sensors J., vol. 15, no. 7, pp. 3781–3792, Jul. 2015.

[4] J. M. Pak, C. K. Ahn, P. Shi, Y. S. Shmaliy, and M. T. Lim, ‘‘Distributed hybrid particle/FIR filtering for mitigating NLOS effects in TOA based localization using wireless sensor networks,’’ IEEE Trans. Ind. Electron., vol. 64, no. 6, pp. 5182–5191, 2017.

[5] T.-J. Ho and Z.-H. Chen, “Mobile location estimation in urban areas using fuzzy-tuned IMM algorithm,” in Proc. IEEE Int. Conf. WCSP, Huangshan, China, Oct. 2012.

[6] T.-J. Ho, B.-Y. Chen, “Robust urban mobile localization using data fusion and intelligent estimation,” in Proc. IEEE Wireless and Mobile Computing, Networking and Communication, Lyon, France, pp.653-658, Oct. 2013.

國際替代計量

一種強健擴展型卡爾曼濾波為基礎的交互多模型算法用於改善在隨機非直視環境下定位性能

全文下載

主題瀏覽