4G Self-Configurable System Based on Nueral Networks

This work present the study on two types of neuronal networks for a fourth generation (4G) self-configurable wireless communications system. Simulations were made in MATLAB to study the behavior of two types of neuronal networks, back propagation network (BPN) and radial basis function (RBF) trained to discriminate signals taken from three systems of wireless communication of second (2G) and third generation (3G). When the results are observed it seems that designed network BPN, generates a smaller error of recognition in addition of which it contains a smaller number of neurons than network RBF, this reduces the system time response in addition of which it allows a saving in the memory use