【摘要】 针对阵列信号处理的特点 ,提出了一种基于径向基函数网络 (RBFNN)的盲波束形成方法。该方法把高阶累积量可消除高斯噪声干扰这一特性和神经网络可并行计算的结构优势有机地结合起来 ,利用高阶累积量估计阵列的导向矢量 ,引入径向基函数网络逼近波束形成的权矢量 ,实现盲波束形成。理论分析和仿真实验表明 ,提出的算法能有效地提高运行速度 ,对系统误差也具有很强的鲁棒性。更多还原

【Abstract】 In this paper a blind beamforming algorithm based on higher order cumulant and radial basis function neural network (RBFNN) is presented according to the characteristic of the array signal. This method combines the characteristic that higher order cumulant can eliminate the effect of Gaussian noise with the structural advantage that the neural network can operate in parallel. Steering vector is estimated by higher order cumulant, and weight vector is approximated by RBFNN to carry out the optimum blind beamforming. This method can not only make the computation rapid but also is robust to system error. Theoretical analysis and computer simulation prove the correctness of this algorithm.更多还原