Artificial neural network based macromodeling approach for two-port RF MEMS resonating structures

Abstract

In this paper, we propose an efficient approach for analysis, design, and optimization of two-port radio frequency microelectromechanical systems (RF MEMS) resonating structures. Methodology utilizes finite element method (FEM) for the prediction of electromechanical responses and fast/accurate mapping with an artificial neural networks (ANNs) technique, toward a final goal - a generic macromodel compatible with modern circuit computer aided design (CAD) tools. Thus, instead of using memory and time demanding full-wave analysis or more extensive and expensive design process using multiple fabrication cycles, a simple yet accurate circuit simulator compatible modeling and optimization procedure is developed.