سعید رضا استاد زاده

In this paper, with the use of radial basis functions of neural networks (RBF-NN), closed-forms for analysis of a typical dual-band Wilkinson power divider is presented. At first, a number of samples (input-output pairs) for characteristic impedances of microstrip lines and isolation resistor in Wilkinson power divider are numerically computed using least square technique and they are then used in the training process of the RBF-NN. After the process is converged, closed-forms for characteristic impedances of microstrip lines and isolation resistor are proposed from which design parameters including length and width of microstrip lines as well as the isolation resistor are efficiently extracted.