神经网络用于相关噪声的自适应抵消

本文提出了一种在自适应噪声抵消器中应用模拟神经网络计算自适应线性滤波器权值的方法，权值的计算时间随线性滤波器的阶数的增加而减小。由于神经网络的实时处理能力。该网络可以用于愉速的噪声抵消。当噪声的自相关时间较线性滤波抽头的总地时间为小时时，此时的神经网络相当于一细胞神经网络，这就大大简化了该网络ＶＬＳＩ的实施。本文最后给出了实例模拟。结果令人十分满意。

Neural Network for Correlative Noise Canceller

In this paper, the realization of adaptive noise cancellation using an analogue neural network is discussed. The network is used to compute the coefficients of a linear transversal filter. The settling time decreases as the filter order increases. Owing to the real-time processing capabilities, the network can be used for fast adaptive noise canceller. The special properties of the tap input correlation matrix result in a cellular network architecture which greatly simplifies the VLSI implementation. Simulation results are given which point out very satisfactory performance.