通信卫星干扰定位的RBF神经网络方法

本文提出了结合卫星多波束天线，利用径向基函数（ＲＢＦ）神经网络实现通信卫星干扰源精确定位的方法，这种方法可获得很高的定位精度，且能直接获得信号ＤＯＡ估计的闭式解，从而避免其它经典的高精度ＤＯＡ估计方法（如ＭＵＳＩＣ、ＭＬ等）所必需的全方位峰值搜索，同时由于神经网络优异的并行运算能力，所提方法具有实时估计的优越性，有望应用于实际的实时定位系统中。计算机仿真也验证了这一方法的可行性。

Satellite interference location based RBF neural network method

This paper presents a new interference location method based on Radial Basis Function (RBF)neural network in conjunction with MBA.The proposed method can directly obtain high accurate closed form solution of DOA estimates.Therefore,it can avoid the all direction peak values searching,which is necessary for the classic high accurate DOA estimation method.Furthermore,the method is computationally effective under real time condition owing to its superior ability of parallel processing,which makes it feasible to carry out in practical interference location system.Simulation results show the effectiveness of the proposed method.