Novel design techniques for miniature circularly-polarized antennas using genetic algorithms

We introduce new methodologies employing a genetic algorithm (GA) to evolve classes of antenna shapes that are circularly polarized and offer optimal performance characteristics such as input impedance, VSWR, bandwidth, and/or gain. These new methodologies are also capable of producing designs that are significantly smaller when compared to a conventional crossed-dipole antenna. It is demonstrated that, by utilizing these design approaches for miniature, circularly polarized, stochastic, crossed-dipole antennas, size reductions of over 40% are typically achieved, and in some cases, it is even possible to realize designs with size reductions up to 86%. An important feature of these antennas is that the size reduction achieved is in terms of area, which differs from the case of stochastic linear dipoles where a reduction in length is achieved at the expense of an increase in width.