Applications of ANN and ANFIS to Predict the Resonant Frequency of L-Shaped Compact Microstrip Antennas

Since the Compact Microstrip Antennas (CMAs) with various shapes are crucial for mobile communication, they take much attention in present days and studies related to analysis and design on them have been increasing day by day. In this work, simple approaches based on Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) for computing the resonant frequency of L-shaped CMAs operating at UHF band have been presented. In order to train and test the ANN and ANFIS models, 192 LCMAs having different physical dimensions and relative dielectric constants were simulated by electromagnetic simulation software named IE3D (TM), which is based on Method of Moment (MoM). 172 of LCMAs were employed for training, while the remainders were utilized for testing the models. Average Percentage Errors (APEs) for training were obtained as 0.345% and 0.090% for ANN and ANFIS models, respectively. The constructed models were then tested over the test data and APEs values were achieved as 0.537% for ANN and 0.454% for ANFIS. Afterwards, the accuracy and validity of ANN and ANFIS models proposed in this work were verified on measurement data of the fabricated LCMAs. The results indicate that ANN and ANFIS can be successfully used to predict the resonant frequency of LCMAs without necessitating any other sophisticated calculations.