Evolutionary Computing based Neuron-Computational Model for Microstrip Patch Antenna Design Optimization

Abstract

In this paper, a novel Evolutionary Computing named Adaptive Genetic Algorithm (AGA) based ANN model is developed for rectangular MPA (Microstrip patch antenna). Considering at-hand and Nextgeneration Ultra wideband application demands, the emphasis has been made on retaining optimal lowcost design with desired cut-off frequency. The proposed method employs multiple sets of theoreticallydriven training instances or patch antenna design parameters which have been processed for normalization and sub-sampling to achieve a justifiable and reliable sample size for further design parameter prediction. Procedurally, the input design parameters were processed for normalization followed by sub-sampling to give rise to a sufficient set of inputs to perform knowledge-driven (designparameter) prediction. Considering limitations of the major at-hand machine learning methods which often undergo local minima and convergence while training, we designed a state-of-art new Adaptive Genetic Algorithm based neuro-computing model (AGA-ANN), which helped to predict the set of optimal design parameters for rectangular microstrip patch antenna. The predicted patch antenna length and width values were later used for verification which achieved the expected frequency. The depth analysis revealed that a rectangular patch antenna with width 14.78 mm, length 11.08mm, feed-line 50 Ω can achieve the cut-off frequency of 8.273 GHz, which can be of great significance for numerous UWB applications.

Hybrid Evolutionary Computing Assisted Irregular-Shaped Patch Antenna Design for Wide Band Applications

“…Recalling the suggestions in [44][45][46][47][48][49][50], which stated that the convergence of PSO can be accomplished only when the particle converges to its local attractor (for 1 ≤ d ≤ D), LA t i =(la t i1 ,…, la t id ,…, la t iD ). It can be denoted mathematically in the equations (15)(16).…”

“…(15) (16) Where the parameter t signifies the number of iterations, while j presents the arbitrary number distributed uniformly over (0, 1), that is φ~ U(0,1). Here, pbest i presents the best earlier position observed by the i th paticle, while the current global solution i.e.…”

“…During the past few years, artificial neural network (ANN) or data-driven machine learning (ML) based approaches have been proposed to estimate a suitable set of design parameters for MPA design. Some of the key methods such as neuro-computing concepts [14][15][16][17][18][19][20][21][22][23][24][25][26] performed design parameter estimation; however, their optimality towards a robust solution remained an unexplored story. This is because most of the existing systems focused on design optimization by learning a set of input patterns, irrespective of the end outcomes and its realistic patterns or performance for WB application.…”