The extended use of microelectromechanical systems (MEMS) in the development of new microinstrumentation for aerospatial applications, which combine extreme sensitivity, accuracy and compactness, introduced the need to simplify their design process in order to reduce the design time and cost. The recent apparition of analogue and mixed signal extensions of hardware descriptions languages (VHDL-AMS, Verilog-AMS and SystemC-AMS) permits to co-simulate the HDL (VHDL and Verilog) design models for the digital signal processing and communication circuitry with behavioral models for the non digital parts (analog and mixed signal processing, RF circuitry and MEMS components). Since the beginning of the microinstrumentation design process the modeling and simulation could help to define better the specifications and in the architecture selection and in the SoC design process in a more realistic environment. We will present our experience in the application of these languages in the design of microinstruments by using behavioral modeling of MEMS.
Proc. SPIE. 5836, Smart Sensors, Actuators, and MEMS II
KEYWORDS: Mathematical modeling, Microelectromechanical systems, Oscillators, Sensors, Linear filtering, Signal processing, Analog electronics, Integrated circuit design, Smart sensors, Systems modeling
The objective of this work is to develop a modeling of a complete smart sensor to be used in a distributed architecture, with the new modeling language, VHDL-AMS. This smart sensor is composed by a sensor or actuator, for example we have used a piezoresistive accelerometer, its signal conditioning module, with both analogue and digital elements, and a bus driver that allows communication with the instrument control device and other sensors. In that way, it is also possible to introduce these microsensors in a distributed architecture that permits communication between microinstruments. This example of modeling through VHDL-AMS shows how this language allows a multitechnological description of a microsystem, including not only electrical signals, but also thermical, kinematic, fluidic, etc. signals. This language also permits to describe systems in different levels of complexity and abstraction, giving the possibility of covering several models from a physical model until a behavioral model, which can be used to obtain a design methodology for MEMS, analogous to the existent design methodology for integrated circuits. The combination of smart sensors models at behavioral level with the microinstrument control circuit models is a first step in the development of a complete design methodology.