Impacts of Machine Learning on Counterfeit IC Detection and Avoidance Techniques

Omid Aramoon1 and Gang Qu2
1University of Maryland, 2Univ. of Maryland, College Park


Globalization of integrated circuit (IC) sup- ply chain has made counterfeiting a major source of concern in the semiconductor industry. To address this concern, extensive e orts have been put into developing e ective counterfeit detection and avoidance techniques. In the re- cent years, machine learning (ML) algorithms have played an important role in development and evaluation of many emerging countermeasures against counterfeiting. In this paper, we aim to investigate impacts of such algorithms on the landscape of anti-counterfeiting schemes. We pro- vide a comprehensive review of prior arts that deploy ma- chine learning to develop or attack counterfeit detection and avoidance techniques. We also discuss future direc- tions for application of machine learning in anti-counterfeit schemes.