Novel counterfeit detection of integrated circuits via infrared analysis: A case study based on the Intel Cyclone II FPGAs
Preston D. Frazier, E. Thomas Gilmore, Isaac J. Collins, Mohamed Chouikha · 2016
The counterfeiting of electronic parts is a critical issue that places the reliability and security of numerous systems, such as financial, defense, and aerospace, in peril. The global cost of counterfeiting is as much as $500 billion (USD) annually and negatively impacts the brand of the world's leading integrated circuit manufacturers. Field Programmable Gate Arrays (FPGAs) rank third of the most fraudulently discovered electronic part. This fact is significant since reconfigurable FPGAs are practically ubiquitous in many critical systems and they possess a high resell value. A noninvasive low-cost capability to detect counterfeit FPGAs within these aforementioned industries' supply chains is imperative to the operational certainty of their respective systems. The authors proposed a novel approach to ascertain inauthentic FPGAs, in particular Intel [formerly Altera] Cyclone II devices though use of infrared (IR) analysis and machine learning.