Development of efficient fault diagnosable design methodologies for analog/mixed-signal integrated circuits

More and more mixed-signal devices are being designed recently for the applications of multimedia, wireless communication, and portable data systems. The analog circuit technology conventionally employed for such applications has been gradually switched to analog/digital mixed-signal circuit technology. Even though much more complicated digital circuits have been widely used in the DSP-based mixed-signal IC, analog circuits will remain for processing or interfacing analog signals. Integrating both digital and analog on a single chip has improved performance and reduced board size and cost. However, the increasing complexity of mixed-signal circuits drastically reduces the controllability and observability of the circuit on the chip. As a result, testing and fault diagnosis of such complex circuits becomes very difficult and expensive. Therefore, the goal of the thesis study is to develop efficient diagnosable design methodologies for analog/mixed-signal integrated circuits and further integrate the methodologies for the development of an automated diagnostic test system.; This study develops a framework of the automated diagnostic test system for analog circuits, including two major analysis tools Diagnosability Analysis and Redesignability Analysis, and Diagnosable Design Methodologies.; The diagnosability analysis estimates the maximum diagnosability a circuit can achieve and locates sections of a circuit having poor diagnosability for a given set of test points, while redesignability analysis derives the input/output (I/O) relations of the portions having poor diagnosability and reconstructs them for diagnosability enhancement. The developed diagnosable design methodologies include two major processes: Test Points Selection and Diagnostic Test Programs Generation. The former process identifies the number of test points required to achieve the desired diagnosability and their locations, while the latter process automatically generates test programs in analog version hardware description language, HDL-A.; A system, namely ADTS (Automated Diagnostic Test System), is also developed to realize the building blocks of the developed framework. A Graphic User Interface (GUI) environment is developed to allow user to input the circuit description and to conduct the desired analysis or analyses. The system is written in JAVA, an object-oriented language, with a data structure that implements spare matrix techniques for efficient memory usage.