Abstract: This paper presents the modelling of an aperture coupled microstrip antenna (ACMSA) using four different Artificial Neural Network (ANN) techniques, which are Back Propagation Network(BPN), Radial Basis Function Network(RBF),Generalized Regression Neural Network(GRNN) and a hybrid model which is a combination of Probabilistic Neural Network(PNN) and BPN. Firstly, these ANN models have been developed using MATLAB software, thereafter with the use of IE3D software, antenna design has been simulated, and validation performed, to check the accuracy of the developed models. Error percentages have been calculated, by comparing both the simulated and ANN models, for both normalized and un-normalized data sets. Thus the neural network approach eliminates the long, time–consuming process of finding various design parameters, using costly software packages.

Keywords: Aperture Coupled Microstrip Antenna (ACMSA), Artificial Neural Network (ANN), BPN, RBF, GRNN, PNN