Design of circular/triangular patch microstrip antennas using a single neural model

An efficient and quick approach based on artificial neural networks (ANN) is being applied on different patches of microstrip antennas since last one decade. Different scientists have proposed different neural models for analyzing the different types of microstrip patches like rectangular, triangular, square, circular etc. whereas few models have also been proposed for designing rectangular, triangular and square patches. But in the available literature of microstrip antennas with neural networks no single model has been proposed till date firstly for calculating the radius of the circular patch microstrip antenna (CPMSA) and secondly for calculating more than one parameter like radius of the CPMSA and side-length of the equilateral triangular patch microstrip antenna (ETMSA) simultaneously. In this paper authors have proposed a multi-layered perceptron feed forward neural network model for calculating the radius/side-length of the circular patch/triangular patch microstrip antennas simultaneously. The model has also been validated on some mathematically generated datasets that are not included in training or testing of the model. The results obtained by the proposed model are in conformity and very good in agreement with the experimental results given in the literature.