Derivation of accurate tensor data of materials in SAW devices by solving a parameter identification problem

Accurate simulation of SAW devices requires accurate material tensor data. However, literature values are in most cases measured on certain engineered test devices and do often not reflect real world production processes with the necessary accuracy. In this work, a parameter identification method is outlined, which can be used to estimate material parameters from admittance measurements of resonators produced by the desired process. In order to speed up the computation, the gradients necessary for the optimization method (e.g. quasi-Newton) are derived using techniques from optimal control theory. The algorithm has been tested on a set of resonators composed of a substrate, an electrode and a passivation layer.