Analog and Mixed-Signal Test and Fault Diagnosis
Analog and mixed-signal (AMS) test and fault diagnosis plays an essential role in circuit design, device production, and instrumentation maintenance. The benefits include correcting design flaws, reducing time-to-market, increasing manufacture yield, and reducing system cost. Usually, AMS test and fault diagnosis consists of three tasks. The first one is fault detection to check if the CUT is faulty or not, which is usually called test in industry. The second task is fault location to find out where the faulty parameters are inside the faulty circuit. The final task is parameter evaluation to compute how much the faulty parameters are deviated from their nominal values. The difficulties for AMS test and fault diagnosis come from ambiguities, increased complexity, reduced accessibility, lack of effective fault model, and increased test cost.
In this presentation, above problems are explored. A verification method based on ambiguity group locating technique is designed for accurate computing. To decrease complexity and increase accessibility, a decomposition technique is implemented to decompose large scale system into smaller subsystems. An optimum test nodes selection algorithm will be introduced to find out a local minimum set whose efficient and solution accuracy is the best comparing with other reported algorithms. To locate multiple analog catastrophic faults, an analog stuck fault location approach is designed removing repetitive simulation required by traditional approaches. One example of mixed-signal system, sampling voltmeter is utilized for model building and analog simulation in order to test and locate its fault mechanism.
Mr. Dong Liu is a Ph.D. student in Electrical Engineering at Ohio University. His dissertation research is in area of analog and mixed-signal test and fault diagnosis. He hold a BE degree of EE from Tianjin University, China in 1993 and a MS degree of EE from China Academy of Sciences in 1996. He was an engineer in Huafeng Test & Control Technology Co. Ltd, China between 1996 and 1998. Between 1998 and 1999, he was a full-time research student working on "Testing Strategies for mixed-signal systems with embedded software" at Electronics and Electrical Engineering Lab, National Institute of Standards and Technology (NIST), Gaithersburg, MD. In 2002, he was nominated by EECS Department for Ohio University Outstanding Graduate Student (Doctoral) Award.