Design Annular Ring Microstrip Antenna Based on Artificial Neural Network

This paper present the design of annular ring microstrip antenna (ARMAs) based on artificial neural network (ANN). Different size of ARMAs have been designed with varied dimensions. The electrical parameters and various dimensions that are obtained for the different resonant frequencies that covering application of GSM, LTE, WLAN and WiMAX band. The employed ANN, computing the resonant frequency of ARMAs. Based on backpropagation (BP) technique, Multi-layer perceptron (MLP) model has been used to trained and test the resonant frequency of ARMAs. By employing BP-(MLP), ANN used 70 ARMAs for training and 10 for test, which selected randomly. The different learning algorithms for training and testing are employed, to achieve the best result. The performance, in which the best result, obtained from Levenberg Marquardt (LM) as compared to other algorithms. The ANN model with LM algorithm, successfully use the resonant frequency of ARMAs, which give best accuracy.