Reliability Improvement on SRAM Physical Unclonable Function (PUF) Using an 8T Cell in 28 nm FDSOI

Su, Ze-Xin; Li, Bo; Zhang, Wei‐Dong; Gao, Jiantou; Su, Xiaohui; Zhang, Gang; Ren, H.; Lü, Pei; Liu, Fanyu; Zhao, Fukun; Shi, Li

doi:10.1109/tns.2021.3126587

Cited by 7 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For uniqueness, PUFs and RNGs require high uniqueness, 0.5 of the Hamming distance. For example, according to a paper, 26) the measured HD of the SRAM array is approximately 0.485-0.497. Although SRAM array is used for PUFs and RNGs, the quality of the randomness requires improvement.…”

Section: Proposed Arwgmentioning

confidence: 99%

“…The proposed ARWG is discussed and compared with physically unclonable functions (PUFs) and RNGs in terms of their use, function, and randomness. [23][24][25][26][27][28][29][30] The randomness of random weights generated by the proposed ARWG is analyzed with device and circuit variations. Furthermore, the requirements of random weights from RC and random weight SNN are evaluated by Hamming distance and Hamming weight, which are indicators of repeatability and uniqueness of the randomness.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dual integration design of approximate random weight generator and computation-in-memory for event-based neuromorphic computing

Misawa,

Koshino,

Liu

et al. 2024

Jpn. J. Appl. Phys.

View full text Add to dashboard Cite

This paper comprehensively analyses dual integration of approximate random weight generator (ARWG) and computation-in-memory for event-based neuromorphic computing. ARWG can generate approximate random weights and perform multiply-accumulate (MAC) operation for reservoir computing (RC) and random weight spiking neural network (SNN). Because of using device variation to generate random weights, ARWG does not require any random number generators (RNGs). Because RC and random weight SNN allow approximate randomness, ARWG only needs to generate approximate random weights, which does not require error-correcting code to correct weights to make the randomness accurate. Moreover, ARWG has a read port for MAC operation. In this paper, the randomness of random weights generated by the proposed ARWG is evaluated by Hamming distance and Hamming weight. As a result, this paper reveals that the randomness required for ARWG is much lower than that for physically unclonable functions and RNGs, and thus the proposed ARWG achieves high recognition accuracy.

show abstract

Section: Proposed Arwgmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Dual integration design of approximate random weight generator and computation-in-memory for event-based neuromorphic computing

Misawa,

Koshino,

Liu

et al. 2024

Jpn. J. Appl. Phys.

View full text Add to dashboard Cite

show abstract

Reliability analysis and comparison of ring-PUF based on probabilistic models

Bian,

Huang,

Lin

et al. 2024

Microelectronics Journal

View full text Add to dashboard Cite

Heavy ion-induced microdose effects on the reliability of planar and FinFET-based SRAM physical unclonable functions

Shao,

Song,

et al. 2024

Applied Physics Letters

View full text Add to dashboard Cite

The static random-access memory (SRAM) physical unclonable function (PUF) uses the power-up states of stored values to derive the identification code. Its lightweight circuit design makes it suitable for power-sensitive satellites. However, the SRAM PUF is significantly affected by high-energy particles and cosmic rays in space, which causes incorrect output of identity authentication codes and further leading to a degradation in reliability. Recently, the main research works are concerned about the impact of cosmic rays induced total ionizing dose effects on the SRAM PUF, and few reports discuss the high-energy particles induced single event effects on the SRAM PUF. This paper presents the reliability results of planar and FinFET-based SRAM PUFs after high-energy heavy ion experiments. Experimental results indicate that 3%–10% of the SRAM PUF bits change from their original power-on states, which demonstrate that ion-induced microdose effects degrade the reliability of SRAM PUFs. In addition, the three-dimensional technology computer-aided design simulation tool is used to analyze the physical mechanisms of the experimental phenomena. Simulation results show that the heavy ion-induced microdose effect ionizes and produces trapped charges at the silicon/insulation interface, which causes a slight change in transistor leakage current or a slight drift in threshold voltage. The physical mechanisms lead to an off-state current mismatch in the SRAM cell and finally affect the stored values after power-on.

show abstract

Reliability Improvement on SRAM Physical Unclonable Function (PUF) Using an 8T Cell in 28 nm FDSOI

Cited by 7 publications

References 27 publications

Dual integration design of approximate random weight generator and computation-in-memory for event-based neuromorphic computing

Dual integration design of approximate random weight generator and computation-in-memory for event-based neuromorphic computing

Reliability analysis and comparison of ring-PUF based on probabilistic models

Heavy ion-induced microdose effects on the reliability of planar and FinFET-based SRAM physical unclonable functions

Contact Info

Product

Resources

About