| Summary: | Online test is a technology that has great importance and value in many applications. It provides unique benefit of real-time condition monitoring, based on which early warning of system failure or degradation and potentially system self-repair can be achieved to avoid more significant impact. In the context of on-line test, embedded off-line test is still a challenging issue for MEMS device, mainly due to their micro scale and what is typically multi energy domain functionality. There are very limited methods to observe and characterize the MEMS components under test, especially for the dynamic behavior. It is also in general more difficult to verify their performance under real working conditions. This work attempts to deliver real online test solutions and methodologies which can be adapted to wide range of MEMS devices. Novel techniques of Bias-Superposition and Multi-Mode Sensing are conceived and engineered with due consideration of the availability of on board resources (e.g. signal processing power) and design/cost overhead. The proposed techniques are demonstrated on two existing MEMS systems: a capacitive MEMS accelerometer and a resistive MEMS based conductance sensor. The concepts are further studied through the design of a piezoresistive multi-functional humidity/pressure sensor. By developing the theoretical model and applying the proposed methodologies on these different structures and applications, this work pushed the State-of-the-Art of micro & nano system's online test in three areas: firstly the work potentially enabled many existing electrical induced off-line test so lutions for MEMS systems to be transfonned to on-line; secondly an on-line test solution for electro-chemical microsystems has been conceived which by traditional methods is difficu lt to implement; last but not least the work advanced from on-l ine test to fault-to lerance within NJEMS sensor systems through resilience of the sensor network.
|
|---|
