"New Silicon Chip 'Fingerprint' for Stronger Hardware Security at Low Cost"

Researchers from NUS (National University of Singapore) Electrical and Computer Engineering have developed new self-healing and self-concealing Physically Unclonable Functions (PUFs). PUFs use the inherent, unique manufacturing variations in microchips to provide digital fingerprints for devices that can authenticate and secure the devices. This technology also helps to prevent the unlawful cloning of hardware, chip counterfeiting, and physical attacks. However, PUFs still face challenges in regard to stability and occasional fingerprint misidentification. The PUFs created by the researchers address these problems with a novel technique in which on-chip sensors and machine learning algorithms are used to predict and identify PUF instability, as well as make the appropriate adjustments to generate PUF output of higher security and stability. Unusual environmental conditions such as temperature, voltage, and noise that often exploited in the execution of physical attacks can be detected using the technique. This article continues to discuss the NUS research team's newest breakthrough in PUF technology and how this achievement significantly improves hardware security. 

NUS News report "New Silicon Chip 'Fingerprint' for Stronger Hardware Security at Low Cost"