NLOS环境下基于Bootstrap方法的改进TOA定位算法研究

Research on Improved TOA Localization Algorithm Based on Bootstrap Method in NLOS Environment

复杂的非视距传播(NLOS)环境带来的误差会严重影响定位算法的精度和稳健性,单纯的迭代优化算法面对大量的NLOS误差表现得并不理想。针对经典的TOA算法对非视距传播误差的敏感性问题,引入Bootstrap抽样及蒙特卡罗思想,提出了一种改进的TOA算法。利用随机化的思想对误差分布进行更加精确的识别,并通过随机模拟还原直线传播数据,结合牛顿迭代法对移动终端进行精确定位。实证分析证明,这种改进的TOA算法具有更高的精确性和稳健性。

The error caused by the complex non line of sight（NLOS）environment seriously affects the accuracy and robustness of the localization algorithm,the results only obtained by iterative optimization algorithm are not satisfactory faced with a large number of NLOS errors.Aiming at the sensitivity of the classical TOA algorithm to NLOS propagation error,an improved TOA algorithm considering of Bootstrap sampling and Monte Carlo is proposed.The idea of randomization is used to identify the distribution of error more accurately,and the linear propagation data of signal are reduced by random simulation,combined with the Newton iterative method for mobile terminal positioning.Finally,the empirical analysis shows that the improved TOA algorithm has higher accuracy and robustness.