AbTech Corporation proposes to apply its unique Statistical Network ModelingTM technology to develop a non contact prognostic assessment tool that 1) acquires, processes, and analyzes electronic radiated signal data, 2) learns the expected characteristics and patterns of specific radiated signals measured from microelectronic semiconductor devices using AbTech's statistical network modeling technology, and 3) applies these learned models to assess and predict the health and time to failure of electronic equipment and identify unexpected leagues and concerning trends in semiconductor performance. The proposed system, called the Abductive Electronics Prognostic Tool (ADEPT), will be able to identify and interpret complex relationships in electronic equipment behavior to substantially reduce the likelihood of incipient failures and semiconductor degradation being undetected until they cause significant damage to critical electronic equipment. ADEPT will automatically analyze expected versus actual values of specified radiated signals, and indications of the significance of the differences. It will automatically learn how to detect and isolate anomalous equipment behavior based on measurable radiated signal data from semiconductor devices. The objectives of this effort are to prototype ADEPT and to demonstrate its exceptional potential value on selected measurable signals provided by the Center for Semicustom Integrated Systems (CSIS) at the University of Virginia.
Keywords: Electronics Health Monitoring Machine Learning Neural Network Prognostics Semiconductor Statistical Network Test
