Related Experiment Video
Updated: Aug 5, 2026

Soft Lithographic Functionalization and Patterning Oxide-free Silicon and Germanium
Published on: December 16, 2011
Multilevel Multimodal Physical Unclonable Functions by Laser Writing of Silicon Carbide Color Centers
Yuxing Ma1, Yue Qin1, Hao Guo1
1State Key Laboratory of Extreme Environment Optoelectronic Dynamic Testing Technology and Instrument, North University of China, Taiyuan 030051, China.
Abstract:
Information security serves as the cornerstone for ensuring the stable development of today's highly digitized era. As cryptographic primitives with high security and robust encryption capabilities, physical unclonable functions (PUFs) are recognized as one of the critical solutions to address information leakage issues. However, the encoding of PUFs often relies on the inherent properties of materials, which limits the potential for further enhancement of their encoding capacity (EC). In this study, we introduce a four-level encoding scheme by leveraging the stochastic characteristics of free radical chemical reactions and energy deposition in the fabrication process of silicon carbide (SiC) color centers. A multilevel multimodal PUF (MMPUF) encoding strategy (ES) for flexible substrates with high EC, low cost, and simple and fast readout was constructed. The spatially random distribution of SiC and silicon vacancy (Vsi) color-center concentrations as well as the offsets of the laser pyrolysis position along the X- and Y-axes are four independent physical properties that ensure the encoding performance of the PUF, achieving a high encoding capacity of 24×10×10 and secure, stable, and unclonable encoding. Furthermore, the integration of the PUF tags with the products through a doping manufacturing process, rather than simple attachment, enhances the security and practicality of the anti-counterfeiting system. The proposed encoding hierarchy based on the offsets provides a novel encoding solution for improving PUF EC.
Related Concept Videos
Carbon Skeletons
Molecular Models

