TY - GEN
T1 - A 2.17-pJ/b 5b-Response Attack-Resistant Strong PUF with Enhanced Statistical Performance
AU - Liu, Kunyang
AU - Li, Gen
AU - Fu, Zihan
AU - Wang, Xuanzhen
AU - Shinohara, Hirofumi
N1 - Funding Information:
This research is supported by ROHM Co., Ltd. and VDEC, The University of Tokyo, in collaboration with Cadence and Mentor Graphics.
Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - This article presents a Strong physically unclonable function (PUF) with a 5-bit response output, which alleviates the critical issue of huge challenge-response pair (CRP) consumption for one authentication. Attack resistance is enhanced not only by improving the substitution-permutation network (SPN) using variable secret look-up tables (LUTs) and a complex permutation XOR box but also by mitigating LUT data collision and coupling Strong PUF functions between adjacent operation rounds to protect an individual round from attack. As a result, response bitstreams pass all NIST SP800-22 randomness tests even using highly correlated challenge inputs, and mainstream modeling attacks with up to 200-Mb training samples cannot achieve higher accuracy than random guess. An automatic data write-back circuit allows the PUF to be fully stabilized by hot carrier injection (HCI) burn-in without exposing sensitive LUT data. The architecture realizes a 1-cycle/round operation and results in 2.17 pJ/b energy and 0.625 bit/cycle throughput.
AB - This article presents a Strong physically unclonable function (PUF) with a 5-bit response output, which alleviates the critical issue of huge challenge-response pair (CRP) consumption for one authentication. Attack resistance is enhanced not only by improving the substitution-permutation network (SPN) using variable secret look-up tables (LUTs) and a complex permutation XOR box but also by mitigating LUT data collision and coupling Strong PUF functions between adjacent operation rounds to protect an individual round from attack. As a result, response bitstreams pass all NIST SP800-22 randomness tests even using highly correlated challenge inputs, and mainstream modeling attacks with up to 200-Mb training samples cannot achieve higher accuracy than random guess. An automatic data write-back circuit allows the PUF to be fully stabilized by hot carrier injection (HCI) burn-in without exposing sensitive LUT data. The architecture realizes a 1-cycle/round operation and results in 2.17 pJ/b energy and 0.625 bit/cycle throughput.
KW - authentication
KW - hardware security
KW - IoT
KW - modeling attack
KW - physically unclonable function (PUF)
UR - http://www.scopus.com/inward/record.url?scp=85141468478&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85141468478&partnerID=8YFLogxK
U2 - 10.1109/ESSCIRC55480.2022.9911472
DO - 10.1109/ESSCIRC55480.2022.9911472
M3 - Conference contribution
AN - SCOPUS:85141468478
T3 - ESSCIRC 2022 - IEEE 48th European Solid State Circuits Conference, Proceedings
SP - 513
EP - 516
BT - ESSCIRC 2022 - IEEE 48th European Solid State Circuits Conference, Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 48th IEEE European Solid State Circuits Conference, ESSCIRC 2022
Y2 - 19 September 2022 through 22 September 2022
ER -