A new adaptation algorithm of adaptive digital filters using neural networks

The technique of adaptive linear filtering is improved. An adaptive filtering procedure that could be used in the application of the neural network techniques is considered and a control system which adjusts parameters of the adaptive filter by means of a multilayered neural network is designed. The neural network can be trained to adaptively produce the parameters by minimizing the error between the desired output and the output signal of the adaptive filter. After training, the neural network´s weights are fixed. Experimental results confirm that the proposed technique has significant effects on the smoothing problem