This paper presents a single-site positioning method based on the joint estimation of propagation time-of-arrival(TOA) and direction-of-arrival(DOA), with the assist of virtual stations in the typical non-line-of-sigh...This paper presents a single-site positioning method based on the joint estimation of propagation time-of-arrival(TOA) and direction-of-arrival(DOA), with the assist of virtual stations in the typical non-line-of-sight(NLOS) environment. Consider the influence of multipath noise on the positioning performance, the proposed method firstly presents a modified high-resolution estimation technique called Multipath noise Limiting Matrix Pencil(MLMP) algorithm to achieve the TOA/DOA estimations, in which the matrix pencil and matrix enhancement process are implemented to deal with the measurements from the uniform linear array(ULA) receiver. Meanwhile, the subspace dimension estimation is improved via an adaptive threshold, for enhancing the performance of high-resolution techniques in low signal-noise-ration(SNR) situation. Next the proposed method generates virtual stations utilizing the known floor plan of surrounding reflectors, and adopts a weighted Least Square(WLS) position estimator to calculate the required position, combining the TOA/DOA estimations with the location of virtual stations. Simulations are conducted to evaluate the proposed method under NLOSconditions, and the results show that comparing with the multipath fingerprinting scheme, the proposed method has better performance in various simulation scenarios.展开更多
基金supported in part by the National Natural Science Foundation of China under Grants numbers 61471164, 61601122, 61741102 and 61571128
文摘This paper presents a single-site positioning method based on the joint estimation of propagation time-of-arrival(TOA) and direction-of-arrival(DOA), with the assist of virtual stations in the typical non-line-of-sight(NLOS) environment. Consider the influence of multipath noise on the positioning performance, the proposed method firstly presents a modified high-resolution estimation technique called Multipath noise Limiting Matrix Pencil(MLMP) algorithm to achieve the TOA/DOA estimations, in which the matrix pencil and matrix enhancement process are implemented to deal with the measurements from the uniform linear array(ULA) receiver. Meanwhile, the subspace dimension estimation is improved via an adaptive threshold, for enhancing the performance of high-resolution techniques in low signal-noise-ration(SNR) situation. Next the proposed method generates virtual stations utilizing the known floor plan of surrounding reflectors, and adopts a weighted Least Square(WLS) position estimator to calculate the required position, combining the TOA/DOA estimations with the location of virtual stations. Simulations are conducted to evaluate the proposed method under NLOSconditions, and the results show that comparing with the multipath fingerprinting scheme, the proposed method has better performance in various simulation scenarios.
