Radial basis function networks in nonlinear signal processing applications

Abstract

We consider radial basis function networks (RBFN) for use in various nonlinear signal processing applications. We first present a simple training algorithm for the RBFN based on stochastic gradients of error. We then demonstrate and discuss the usefulness of RBFNs in various applications including channel equalization, interference cancellation, time-series prediction, and nonlinear filtering.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>