A reconfigurable PUF structure with dual working modes based on entropy separation model

作者：Wang, Yanjie; Liang, Huaguo; Wang, Yuchuan; Yao, Liang; Yi, Maoxiang; Huang, Zhengfeng; Lu, Yingchun^*

来源：MICROELECTRONICS JOURNAL, 2022, 124: 105445.

DOI：10.1016/j.mejo.2022.105445

• FPGA
• Physical unclonable PUF)
• True random number generator (TRNG)
• Reconfigurable
• Separation model

Privacy Protection Mutual Authentication (PPMA) in the Internet of Things uses physical unclonable functions (PUFs) and true random number generators (TRNGs) as security primitives, providing a lightweight solution for key generation and protection. At present, the designs for achieving PUF and TRNG simultaneously on field programmable gate array (FPGA) lack theoretical support, and have poor randomness and stability. Therefore, this paper provides a separation model of PUF entropy (manufacturing delay) and TRNG entropy (jitter delay) based on ring oscillators (ROs). Based on this model, a reconfigurable PUF structure is proposed, which can be reconstructed to a TRNG by changing the working mode. The proposed structure not only improves the area utilization, but also realizes the low-power consumption. Experiments on Xilinx Spartan-6 FPGAs and Xilinx Virtex-6 FPGAs show that the proposed structure has wonderful performance. The PUF has excellent uniqueness (average resolution is 50.21%) and reliability (the average repeatability rate is 96.98%). TRNG's output sequences have high randomness and pass all National Institute of Standards and Technology (NIST) tests (average pass rate is 99%).