Communication channel equalisation using complex-valued minimal radial basis function neural network

Presents a sequential learning algorithm and evaluates its performance by using it to build up an RBF network for complex-valued communication channel equalisation problems. The algorithm is referred to as the complex minimal resource allocation network (CMRAN) algorithm and it is an extension of the MIRAN algorithm originally developed for online learning in real valued RBF networks. CMRAN has the ability to grow and prune the (complex) RBF network's hidden neurons to ensure a parsimonious network structure. Simulation results presented clearly show that CMRAN is very effective in equalisation problems with performance achieved often being superior to that of some of the well-known methods.