ANN based electromagnetic models for the design of RF MEMS switches

Artificial neural network (ANN)-based modeling approach for the design of radio frequency microelectromechanical system (MEMS) switches is presented. In the proposed methodology, a finite element method analysis is performed for the switch characterization and for the creation of training and testing sets for the ANN. Developed full-wave/ANN models allow for circuit level design, simulation, and optimization in a schematic of a circuit simulator. The simulation time is reduced for two orders of magnitude, while the accuracy of the full-wave method is maintained.