中文摘要：

为了减小NLOS传播的影响,提出基于改进RBF网络的TDOA/AOA定位算法.模拟退火算法与k-均值聚类算法相结合的RBF网络训练算法,利用模拟退火算法全局寻优能力改变k-均值算法易陷入局部极值的缺点.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WLS算法和k-均值聚类的RBF网络定位算法.

英文摘要：

In order to mitigate the effect of NLOS propagation, a TDOA/AOA localization algorithm based on the improved RBF neural network is proposed. The methods for training RBF neural network is combined to k- means duster with simulated annealing algorithm, which use the global optimize ability of SA to remedy the local extremum shortcoming of k- means. 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 WLS algorithm and k- means RBF neural network algorithm in NLOS environment.