Automatic Sampling Scheme in Parametric Modeling of Microwave Antennas

This paper introduces a scheme about automatic collection of sampling data in parametric modeling of microwave antennas. The co-simulation of MATLAB and HFSS is used to implement this process. Vector fitting is performed to obtain the coefficients of the transfer function. The nonlinear mapping between geometric parameters of microwave antennas and transfer function coefficients is learned by the artificial neural network. An example of the U-slot microstrip patch antenna is presented to show the advantage of this work.