Random fractal-enabled physical unclonable functions with dynamic AI authentication

This study proposes a novel anti-counterfeiting system using random fractal-network Physical Unclonable Functions (PUFs) and AI-based authentication. A fractal-guided film annealing strategy creates unique, stable Au network-based PUFs with configurable complexity. A dynamic deep learning authentication system, featuring an expandable database, achieves efficient and reliable authentication with zero false positives. The method demonstrates potential for multi-level security through Raman-based chemical encoding. These mass-producible tags offer a competitive solution for high-level anti-counterfeiting and data encryption.

Ningfei Sun, Ziyu Chen, Yanke Wang, Shu Wang, Yong Xie, Qian Liu