Role of Boundary Dynamics in Improving Efficiency of Particle Swarm Optimization on Antenna Problems

Attainment of global optimal solution and reduction of computational time and resources have always been a trade off issue in formulation of nature inspired algorithms, this trade off challenge has brought in a deluge of new algorithms proposing their efficacy over one another. Particle swarm optimization (PSO) algorithm and its variants are quite popular in optimizing antenna designs particularly due to their algorithmic simplicity and fast convergence rates. However, as a matter of the ever present trade off, fast convergence often leads to sub optimal solutions. In some recent researches it has been shown that a class of boundary handling algorithms of PSO, known as position regulated boundary conditions (PRBCs), can minimize the trade off between fast convergence and accomplishment of global optimal solution, here, the performance of these boundary algorithms are compared with that of other established ones over three different optimization targets of antenna design namely, "Multi target optimization in linear antenna arrays", "Inset feed position optimization of rectangular patch antennas" and "Edge feed position optimization of rectangular patch antennas". Results show that the use of PRBCs in PSO leads to impressive improvement in the optimization efficiency in terms of lesser computational time and attainment of global optimal solution.