An ANN-Based Synthesis Model for the Single-Feed Circularly-Polarized Square Microstrip Antenna With Truncated Corners

An artificial neural network-based synthesis model is proposed for the design of single-feed circularly-polarized square microstrip antenna (CPSMA) with truncated corners. To obtain the training data sets, the resonant frequency and Q-factor of square microstrip antennas are calculated by empirical formulae. Then the size of the truncated corners and the operation frequency with the best axial ratio are obtained. Using the Levenberg-Marquardt (LM) algorithm, a three hidden layered network is trained to achieve an accurate synthesis model. At last, the model is validated by comparing its results with the electromagnetic simulation and measurement. It is extremely useful to antenna engineers for directly obtaining patch physical dimensions of the single-feed CPSMA with truncated corners.