A Novel RRAM Based PUF for Anti-Machine Learning Attack and High Reliability

Expand
  • (1. Academy of Electronic Information Engineering, North China University of Technology, Beijing 100144, China; 2. Institute of Microelectronics, Tsinghua University, Beijing 100084, China)

Online published: 2019-01-28

Abstract

Due to the unique response mechanism, physical unclonable function (PUF) has been extensively studied as a hardware security primitive. And compared to other PUFs, the resistive random access memory (RRAM) based PUF has more flexibility with the change of conductive filaments. In this work, we propose an exclusive or (XOR) strong PUF based on the 1Kbit 1-transistor-1-resistor (1T1R) arrays, and unlike the traditional RRAM based strong PUF, the XOR PUF has a stronger anti-machine learning attack ability in our experiments. The reliability of XOR RRAM PUF is determined by the read instability, thermal dependence of RRAM resistance, and aging. We used a split current distribution scheme to make the reliability of XOR PUF significantly improved. After baking for 50 h at a high temperature of 150?C, the intra-chip Hamming distance (Intra-HD) only increased from 0 to 4.5%. The inter-chip Hamming distance (Inter-HD) and uniformity are close to 50% (ideally). And it is proven through the NIST test that XOR PUF has a high uniqueness.

Cite this article

DAI Lan *(戴澜), YAN Qiangqiang (闫强强), YI Shengyu (易盛禹), LIU Wenkai (刘文楷), QIAN He (钱鹤) . A Novel RRAM Based PUF for Anti-Machine Learning Attack and High Reliability[J]. Journal of Shanghai Jiaotong University(Science), 2019 , 24(1) : 101 -106 . DOI: 10.1007/s12204-019-2043-0

References

[1] PAPPU R, RECHT B, TAYLOR J, et al. Physicalone-way functions [J]. Science, 2002, 297(5589): 2026-2030. [2] LU X Y, HONG L Y, SENGUPTA K. An integratedoptical physically unclonable function using processsensitivesub-wavelength photonic crystals in 65 nmCMOS [C]//International Solid-State Circuits Conference.San Francisco, CA, USA: IEEE, 2017: 272-273. [3] BHARGAVA M, CAKIR C, MAI K. Comparison of bistableand delay-based physical unclonable functionsfrom measurements in 65 nm bulk CMOS [C]//CustomIntegrated Circuits Conference. San Jose, CA, USA:IEEE, 2012: 1-4. [4] MAES R, TUYLS P, VERBAUWHEDE I. A softdecision helper data algorithm for SRAM PUFs[C]//International Symposium on Information Theory.Seoul, Korea: IEEE, 2009: 2101-2105. [5] LIU R, WU H Q, PANG Y C, et al. A highly reliableand tamper-resistant RRAM PUF: Design and experimentalvalidation [C]//International Symposium onHardware Oriented Security and Trust. McLean, VA,USA: IEEE, 2016: 13-18. [6] R¨UHRMAIR U, SEHNKE F, S¨OLTER J. Modeling attackson physical unclonable functions [C]//17th ACMConference on Computer and Communications Security.Chicago, IL, USA: ACM, 2010: 237-249. [7] PANG Y C, WU H Q, GAO B, et al. A novel PUFagainst machine learning attack: Implementation ona 16Mb RRAM chip [C]//International Electron DevicesMeeting. San Francisco, CA, USA: IEEE, 2017:12.2.1-12.2.4. [8] LIU R, WU H Q, PANG Y C, et al. Experimental characterizationof physical unclonable function based on1 kb resistive random access memory arrays [J]. IEEEElectron Device Letters, 2015, 36(12): 1380-1383. [9] PANG Y C, WU H Q, GAO B, et al. Design andoptimization of strong physical unclonable function(PUF) based on RRAM array [C]//International Symposiumon VLSI Technology, Systems and Application.Hsinchu, China: IEEE, 2017: 1-2. [10] PANG Y C, WU H Q, GAO B, et al, Optimizationof RRAM-based physical unclonable function with anovel differential read-out method [J]. IEEE ElectronDevice Letters, 2017, 38(2): 168-171.
Outlines

/