Multi-Objective Evolutionary Optimization of Aperiodic Symmetrical Linear Arrays

Abstract

In this paper, a multi-objective approach is applied to the design of aperiodic linear arrays of antennas. The adopted procedure is based on a standard Matlab implementation of the Controlled Elitist Non-Dominated Sorting Genetic Algorithm II. Broadside symmetrical arrays of isotropic radiators are considered with both uniform and non-uniform excitations. The work focuses on whether, and in which design conditions, the aperiodic solutions obtained by the adopted standard multi-objective evolutionary procedure can approximate or outperform the Pareto-optimal front for the uniformspacing case computable by the Dolph-Chebyshev method.