伪随机扩频系统中抑制窄带干扰的一种神经网络方法

本文研究了伪随机扩频通信系统中抑制窄带干扰的一种神经网络方法，这种方法利用神经网络实时快速最佳地计算作为干扰抑制器的横向滤波器权系数，从而快速实时跟踪窄带干扰并抑制它，比传统的LMS样度下降法以及递归最小平方算法更为优越，理论分析和计算机模拟都表明所提出的神经网络方法在伪随机扩频系统中抑制窄带干扰的有效性。

Neural Network Approach for Rejecting Narrow-Band Interferences in PN Spread Spectrum Systems

A neural network approach for rejecting narrow band interferences in pseudorandom noise spread spectrum systems in investigated in this paper. The basic principle is that, the interference can be tracked and suppressed quickly and in real time, by using the neural network to compute the optimum tap coeffcients of thetransversal filter, being the interfenence suppressor, quickly, in real time, and robustly. This neural network approach have superior performance than traditional least mean square (LMS) algorithms and recursive least square(RLS)algorithms, and is effective and feasible in suppressing the narrow band interferences in PN spread spectrum systems.