Particle swarm optimization for the design of square loop frequency selective surfaces considering a model of dielectric effective permittivity Mario A. Rodriguez Barrera Universidad Distrital, Bogotá, Colombia, and Walter Pereira Carpes Jr Department of Electrical Engineering, GRUCAD/EEL/CTC/UFSC, Florianópolis, Brazil Abstract Purpose – The purpose of this paper is to present the results of a particle swarm optimization (PSO) method applied in the design of a square-loop frequency selective surface (FSS) via the equivalent circuit model (ECM), considering the dielectric effective permittivity as a variable in the optimization problem. Design/methodology/approach – In the optimization process considered, besides the FSS square loop geometric parameters, the thickness and relative permittivity of dielectric material used as support are included as variables in the search space, using for this a model of dielectric effective permittivity introduced by the authors in a previous work. Findings – Square loops were designed and the obtained results were compared with designs reported in literature for applications in wireless local area network and long-term evolution 4G systems. The low computational cost is remarkable as well as the acceptable accuracy obtained with the proposed approach. The PSO method results were implemented with the ECM and compared with those obtained via Ansys – high frequency structure simulator commercial software simulations. Originality/value – The lack of a model of dielectric effective permittivity for the ECM causes a restricted search space in the stochastic FSS design process limited to only geometric parameters, as it is reported in the available literature. The proposed approach simplifies and makes more flexible the design process, and allows guiding the FSS design to unit cell surface and/or dielectric thickness of small dimensions. Keywords Particle swarm optimization, Dielectric effective permittivity, Equivalent circuit model, Frequency selective surfaces Paper type Research paper 1. Introduction Frequency selective surfaces (FSS) are spatial electromagnetic filters that, due to its electromagnetic behavior, are used to control the propagation of electromagnetic energy in several scientific, industrial and military areas (Balanis, 2011; Wu, 1995; Munk, 2005) and, in recent years, in wireless communication applications (Raspopoulos and Stavrou, 2011; Subrt and Pechac, 2012; Li et al., 2011; Subrt et al., 2010; Valdez and Campos, 2014). Different numerical electromagnetic methods can be used to analyze the electromagnetic behavior of FSS, as the finite element method, moments method or the finite difference time domain method. The main disadvantage of this methods is the COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering Vol. 35 No. 5, 2016 pp. 1643-1655 © Emerald Group Publishing Limited 0332-1649 DOI 10.1108/COMPEL-10-2015-0362 Received 6 October 2015 Revised 16 February 2016 Accepted 24 March 2016 The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0332-1649.htm 1643 Particle swarm optimization