Abstract: The mine environment is complex and variable,with various interference factors such as non-line-of-sight (NLOS) propagation. Traditional positioning methods are difficult to meet the requirements for high-precision positioning and have seriously affected underground safety production and management. By combining 5G technique with the Chan-Taylor algorithm,a new method for mine personnel positioning is proposed. Firstly,for data without NLOS errors,the Chan-Taylor positioning algorithm is improved and combined with the residual weighting strategy to enhance the accuracy and stability of positioning. Secondly,for the influence of NLOS errors,a positioning algorithm integrating the Chan algorithm and particle filtering is proposed,which effectively handles the nonlinear positioning problem and further improves the positioning accuracy. Finally, taking a certain mine as an example,the proposed method was tested. The results show that the improved Chan-Taylor positioning algorithm has higher positioning accuracy and stability in single-target and dual-target simulation positioning compared to traditional algorithms. The positioning algorithm integrating the Chan algorithm and particle filtering has an average error smaller than the comparison algorithm under different trajectories and NLOS error conditions,and has higher positioning accuracy. The positioning model combining 5G technique for underground personnel positioning has a higher accuracy of 0. 2 ~ 0. 9 m, which can effectively meet the requirements of underground personnel positioning accuracy and stability for the intelligent construction of mines. 

Key words: underground positioning,non-line-of-sight error,5G technique,Chan-Taylor algorithm