研究成果

Modern integrated circuits are increasingly produced through outsourced and globally distributed supply chains, exposing hardware intellectual property to piracy, reverse engineering, overproduction, counterfeiting, and hardware Trojans. Logic locking has emerged as a major countermeasure by making correct circuit functionality dependent on a secret key. However, despite extensive research on SAT-resilient, approximation-resilient, removal-resilient, and structurally hardened schemes, many practical locking methods remain evaluated mainly against specific attacks rather than through general security definitions. This survey examines logic locking through the lens of program obfuscation, comparing theory-driven approaches, which provide rigorous notions of security but suffer from prohibitive hardware overhead, and implementation-driven approaches, which achieve practical efficiency but often expose recognizable structural artifacts. Through this comparison, we identify a central gap between cryptographic security guarantees and ASIC-oriented implementation constraints. Finally, we argue that future logic locking research requires hardware-specific obfuscation definitions and leakage-aware security metrics that can bridge theoretical rigor and practical efficiency.