On generalized adaptive neural filters with application to enhancing EKG signals

Abstract

A class of nonlinear adaptive filters called generalized adaptive neural filters (GANFs) developed to generalize stack filters to a larger class of nonlinear filters and to outperform stack filters is discussed. They are based on the theory of stack filters and neural networks. It is shown that GANFs have better noise suppression performance than stack filters. The implementation of GANFs is discussed, and the theoretical analysis in regards to the performance of GANFs on EKG signal processing is also presented.<>