隐蔽空间下超宽带TDOA定位算法研究

针对隧道等隐蔽空间环境复杂,干扰严重,最小二乘位置解算方法定位精度低的情况,研究一种结合最小二乘(Least Square,LS)位置求解方法的BP(Back Propagation)神经网络定位算法。首先基于超宽带(Ultra Wide Band,UWB)系统到达时间差(Time Difference Of Arrival,TDOA)利用最小二乘法进行位置解算,然后根据解算得到的初始位置进行TDOA测量值修正,BP神经网络基于修正后的TDOA对LS定位结果进行优化。仿真结果表明,该方法可以有效克服非视距误差影响,定位精度大幅提高,且在不同误差环境下具有一定通用性。

Research on UWB TDOA Location Algorithm for Indoor,Underground and Tunnel Environments

In a complex environment with serious interference,e.g.in a tunnel,the positioning accuracy of the least square method(LS)is low.In this paper,a BP neural network location algorithm combined with least square is studied.Firstly,the least square method is used to determine the position based on the time difference of arrival(TDOA)in UWB system,and then the measured value of TDOA is modifi ed according to the initial position obtained by the LS method.The modifi ed TDOA is used by the BP neural network to get fi nal location results.The simulation results show that the infl uence of NLOS(non-line-of-sight)is reduced and the accuracy of the proposed LS-BP method is greatly improved compared with that of the LS method.The proposed LS-BP method can be applied to handle errors in certain diff erent environments.