FTAI: Frequency-based Trojan-Activity Identification Method for 3D Integrated Circuits

Three-dimensional (3D) integration emerges as a promising solution to enable technology further scaling. However, 3D chip fabrication and assembly involve more steps than the manufacturing process for conventional 2D integrated circuits (ICs). Recent literature indicates that the extended supply chain for 3D ICs could bring in new security threats, such as 3D hardware Trojans. This work proposes a Frequency-based Trojan-Activity Identification (FTAI) method to detect 3D hardware Trojans. The FTAI method is capable of differentiating the frequency changes induced by Trojans from those caused by process variation noise. Our case study indicates that the proposed peak distance metric is over 30× higher than the Euclidean distance used in the existing literature. Theoretical analysis and simulation results show that the proposed method can tolerate more noise than the time-domain Trojan detection method and thus improve the Trojan detection rate by 38.1%.