System Level Modeling of the Relevant Physical Effects of Inertial Sensors Using Order Reduction Methods

• Published in: Analog integrated circuits and signal processing Vol. 44; no. 2; pp. 129 - 135
• Main Authors: Reitz, Sven, Döring, Christian, Bastian, Jens, Schneider, Peter, Schwarz, Peter, Neul, Reinhard
• Format: Journal Article
• Language: English
• Published: 01.08.2005
• ISSN: 0925-1030; 1573-1979
• DOI: 10.1007/s10470-005-2593-y

Sensor development in the field of microsystem technology is driven by steadily increasing demands on sensor resolution and performance and the need to reduce costs. Thus, interactions between mechanical and electronic components of a complex sensor system become more important and have to be considered in the design process, preferably already in early design steps. That is why a system simulation of the entire sensor system is necessary to be able to verify different design steps and to decrease the number of expensive prototyping cycles. Due to typical combinations of different physical domains (mechanics, electronics, thermodynamics, etc.) a simulation of the entire system based on partial differential equations using e.g. finite element methods (FEM) is often impossible or too expensive concerning computing time. Furthermore, many details of the component behavior calculated by FEM are not needed at system level. A successful approach to system simulation of sensor systems is to derive behavioral models from FEM descriptions and to combine them with system level models of electronics. In the following a method for automatic generation of behavioral models for micromechanical components using order reduction methods will be introduced.