Neurocomputational analysis of a multiband reconfigurable planar antenna

Procedures using neural networks are developed for characterizing multiband reconfigurable antennas. A multilayer perceptron (MLP) is used to locate the operational frequency bands of the antenna at different reconfigured conditions. Another self-organizing map (SOM) neural network accomplishes the task of locating the switches to be turned ON for a desired frequency response. The developed formulation is tested on a laboratory prototype antenna.