基于RBF神经网络的TDOA／AOA定位算法  

TDOA/AOA Localization Algorithm Based on RBF Neural Network

作　　者：毛永毅[1,2] 李明远[3] 张宝军[4] 

机构地区：[1]中国科学院国家授时中心 [2]西安邮电学院电信系,西安710061 [3]西安交通大学电子与信息工程学院 [4]西安邮电学院电信系

出　　处：《计算机工程》2008年第3期52-55,共4页Computer Engineering

基　　金：陕西省自然科学基金资助项目(2004F12)

摘　　要：为了减小NLOS传播的影响,提出基于RBF网络的TDOA/AOA算法。利用RBF神经网络对NLOS传播的误差进行修正,使用TDOA/AOA算法进行定位。仿真结果表明该算法减小了NLOS传播的影响,在NLOS环境下有较高的定位精度,性能优于TDOA/AOA算法、Taylor算法、Chan算法和最小二乘(LS)算法。In order to mitigate the effect of NLOS propagation, a TDOA/AOA localization algorithm based on the RBF neural network is proposed. The RBF neural network is made use of to correct the error of NLOS propagation, then position is calculated by TDOA/AOA algorithm. The simulation results indicate that the effect of NLOS propagation is mitigated by this algorithm. Its location accuracy is significantly improved and the performance of this algorithm is better than that of TDOA/AOA algorithm, Taylor algorithm, Chan algorithm and LS algorithm in NLOS environment.

关 键 词：到达时间差 到达角 非视距传播 神经网络 

分 类 号：TN929.53[电子电信—通信与信息系统]