Predicting the Resonant Frequency of E-shaped Compact Microstrip Antennas by Using Anfis and SVM

Abstract

This paper deals with the application of two robust methods, namely the adaptive neuro-fuzzy inference system (ANFIS) and the support vector machine (SVM), to predict the resonant frequency of E-shaped compact microstrip antennas (ECMAs) operating at UHF band. Firstly, the resonant frequencies of 144 ECMAs with varied dimensions and electrical parameters were simulated with $\hbox {IE3D}^\mathrm{TM}$ based on method of moment (MoM).The ANFIS and SVM models for computing the resonant frequency were then built by considering the simulation data. 130 simulated ECMAs were utilized for training and the remaining 14 ECMAs were used for testing the ANFIS and SVM models. The average percentage errors (APE) regarding the computed resonant frequencies for training by ANFIS and SVM were obtained as 0.105 and 0.387 %, respectively. The constructed models were then tested over the test data and APE values were achieved as 0.361 % for ANFIS and 0.480 % for SVM. Additionally, agreement between ANFIS and SWM results was investigated with $r^{2}$ method. The results of both ANFIS and SWM were compared with those of the methods previously published in the literature, and a very good agreement between the models proposed and the simulations was obtained. Also, the validity and accuracy of these models were verified on the fabricated ECMA operating at 2.4 GHz. The results achieved in this work show us that both ANFIS and SVM models can be used to predict the resonant frequency of ECMAs as the useful and versatile methods without involving any sophisticated methods.