2019/10/16 by Zhen-Yu Liang, Hao-Hsuan Wei, Tsung-Te Liu · 1 citation
Computer Science · Engineering · Neuroscience · #Physical Unclonable Functions (PUFs) and Hardware Security #Integrated Circuits and Semiconductor Failure Analysis #Neuroscience and Neural Engineering
paper · doi:10.1109/jssc.2019.2942374
This article presents a wide-range, variation-resilient physically unclonable function (PUF) to provide a reliable security primitive solution. The proposed ring oscillator-based PUF employs several stabilization techniques to maintain robustness under environmental variations. The proposed PUF design uses a delay cell topology with an adjusted signal slope to amplify the impact of device variability on delay mismatch. Moreover, an online calibration system that extracts in situ PUF delay information and re-configures stage interconnects is employed to further enlarge frequency mismatch and improve stability. Measurement results from a prototype fabricated in a 28-nm CMOS technology show that the proposed PUF design achieves a highly stable performance over a wide range of operating conditions, with a worst case bit error rate (BER) of 0.55% across 0.4-1.3 V and -40 to 125 °C. The measured BER sensitivities to supply voltage (0.0546%/0.1 V) and temperature (0.0052%/10 °C) demonstrate the stability improvements of 2.38 and 28.84 times, respectively, when compared with the state-of-the-art results.