一种基于神经网络的非线性自适应滤波器

提出了基于Ａｄａｌｉｎｅ－ＭＬＰ递归神经网络的非线性自适应滤波器。这种新的滤波器运用了自适应ＩＲ滤波器理论，具有神经网络分布式并行信息处理，较好的容错特性和鲁棒效应等优点，且比传统的滤波器有更好的收敛特性，并可方便地实现非线性滤波。此外，还给出了Ａｄａｌｉｎｅ－ＭＬＰ递归网络在线滤波算法及其在非线性自适应噪声抵消器中的应用。最后进行了计算机仿真。

A Non-Linear Adaptive Filter Based on Neural Network

A non linear adaptive filter based on the Adaline MLP recursive neural network is proposed in this paper.The theory of adaptive IIR filter is applied into the novel filter which is of the advantages of neural networks,such as parallel distributed information processing,fault tolerance and robustness.Compared with traditional structures,the filter has better convergence properties and may realize the non linear filter.Moreover,the paper gives the on line filtering algorithm for neural network filters and the practical application of non linear adaptive noise cancelling system.Computer simulations are included to illustrate the analytical results.