Successive Boolean Optimization of Planar Pixel Antennas

We propose a novel heuristic method for optimizing planar pixel antennas which we refer to as successive exhaustive Boolean optimization (SEBO). The key step in SEBO is to cyclically optimize a multivariable binary optimization problem by exhaustively searching a subset of binary variables. An adaptive version of SEBO is also introduced. We provide both simulation and experimental results for optimized antennas covering single and dual bands for mobile phone applications. Comparisons with genetic algorithms, binary particle swarm optimization, and angle modulate differential evolution are also provided and these show that SEBO and adaptive SEBO have up to 50% less computational complexity and can satisfy the design objective more consistently. A major advantage of SEBO and adaptive SEBO is that only one parameter needs to be set up for the optimization, while for evolutionary optimization two or more parameters usually need to be set up.