Antenna Modeling by Elementary Sources Based on Spherical Waves Translation and Evolutionary Computation

A method based on spherical waves translation and evolutionary computation is used to model antennas efficiently. A set of infinitesimal dipoles is taken into account instead of the antenna, optimizing the number and positions of the dipoles by using a binary genetic algorithm (GA) for modeling the near field with a desired accuracy. It is shown how this approach can model volumetric antennas by a set of equivalent sources placed on a ground plane accurately. Moreover, large finite arrays can be quickly simulated since GA is only applied to analyze an isolated element, avoiding the need of distributing equivalent sources regularly over an equivalent surface.