USING AN RBF NETWORK FOR BLIND EQUALIZATION: DESIGN AND PERFORMANCE EVALUATION

The design of adaptive equalizers is an important topic for practical implementation of ecient digital communica-tions. In this paper, the application of a radial basis func-tion neural network (RBF) for blind channel equalization is analysed. This architecture is well suited for equalization of nite impulse response (FIR) channels partly because the network model closely matches the data model. This al-lows a rather straightforward design of an optimal receiver, in a Bayesian sense. It also provides a simple framework for data classication, in which more complex nonlinear distor-tions can be accomodated with virtually no modications. A clustering algorithm for dynamic creation and combina-tion of local units is proposed, which eliminates the need for channel order estimation. An initialization procedure for the output linear layer is also presented. The network performance is illustrated with Monte Carlo simulations for a family of random channels. 1