Project Grant 2302537

This Project Grant award from the National Science Foundation's Integrative Activities program (CFDA 47.083) supports a project titled "EXCELLENCE IN RESEARCH: A HIERARCHICAL MACHINE LEARNING APPROACH FOR SECURING OF NOC-BASED MPSOCS AGAINST THERMAL ATTACKS" at North Carolina Agricultural and Technical State University (NC A&T).

The $575,955 award will fund the development of a hierarchical machine learning approach to monitor and detect compromised thermal sensors in multi-processor system-on-chips (MPSoCs) that are vulnerable to hardware Trojan attacks. The project aims to implement countermeasures at the network-on-chip (NoC) routers, an on-chip machine learning accelerator, and cloud-based machine learning processing to improve the accuracy of identifying malicious thermal sensor behavior. As a sub-awardee, the University of Mississippi will contribute to the development of the router-level countermeasures, thermal data collection, and on-chip ML accelerator integration. The University of North Carolina at Chapel Hill will also support the project by verifying the router-level countermeasures through undergraduate student research. The project runs from Aug 15, 2023 to Jul 31, 2026.