Utilizing Machine Learning Techniques for Identifying and Diagnosing Integrated Circuit Faults

The findings of the research that was carried out present a method that is driven by machine learning, which may be utilizedto conduct thermal testing of integrated circuits. To carry out this method, the current temperature distribution is compared to a reference distribution that is suitable for the energy state of the system. This comparison is made to ensure that the procedure is carried out smoothly. In addition, as part of this research endeavor, an investigation is being conducted into a methodology for the positioning of temperature sensors. This placement approach is advised for discovering and resolving issues that are associated with integrated circuits (ICs), and it is recommended that it be used. Our strategy for failure localization makes use of the average temperatures in some sub-areas of the integrated circuit, in addition to the interconnections that stand between those temperatures. This allows us to pinpoint the exact location of the failure. The results of the simulations that were performed for the sensor placement strategy, defect detection, and localization are presented in this research. The findings regarding the efficiency of the testing strategy are displayed through the utilization of statistical approaches to display the findings.