Blind equalization using a radial basis function neural network

Nonlinear filters based on neural networks can be used for adaptive signal processing in a wide range of applications, e.g. underwater acoustic communications. In this paper, a radial basis function (RBF) neural network is used for blind adaptive equalization with higher order statistics. The RBF network proposed in this paper has several features which make it a suitable structure for blind equalization, such as a training algorithm where unsupervised learning appears naturally. The resulting structure is modular, and a real-time implementation is feasible using simple hardware. A performance analysis of the network, based on simulated and real data, is presented