TY - GEN
T1 - Side-Channel Analysis on Lattice-Based KEM Using Multi-feature Recognition - The Case Study of Kyber
AU - Ma, Yuan
AU - Yang, Xinyue
AU - Wang, An
AU - Wei, Congming
AU - Chen, Tianyu
AU - Xu, Haotong
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.
PY - 2024
Y1 - 2024
N2 - Kyber, selected as the next-generation standard for key encapsulation mechanism in the third round of the NIST post-quantum cryptography standardization process, has naturally raised concerns regarding its resilience against side-channel analysis and other physical attacks. In this paper, we propose a method for profiling the secret key using multiple features extracted based on a binary plaintext-checking oracle. In addition, we incorporate deep learning into the power analysis attack and propose a convolutional neural network suitable for multi-feature recognition. The experimental results demonstrate that our approach achieves an average key recovery success rate of 64.15% by establishing secret key templates. Compared to single-feature recovery, our approach bypasses the intermediate value recovery process and directly reconstructs the representation of the secret key. Our approach improves the correct key guess rate by 54% compared to single-feature recovery and is robust against invalid attacks caused by errors in single-feature recovery. Our approach was performed against the Kyber768 implementation from pqm4 running on STM32F429 M4-cortex CPU.
AB - Kyber, selected as the next-generation standard for key encapsulation mechanism in the third round of the NIST post-quantum cryptography standardization process, has naturally raised concerns regarding its resilience against side-channel analysis and other physical attacks. In this paper, we propose a method for profiling the secret key using multiple features extracted based on a binary plaintext-checking oracle. In addition, we incorporate deep learning into the power analysis attack and propose a convolutional neural network suitable for multi-feature recognition. The experimental results demonstrate that our approach achieves an average key recovery success rate of 64.15% by establishing secret key templates. Compared to single-feature recovery, our approach bypasses the intermediate value recovery process and directly reconstructs the representation of the secret key. Our approach improves the correct key guess rate by 54% compared to single-feature recovery and is robust against invalid attacks caused by errors in single-feature recovery. Our approach was performed against the Kyber768 implementation from pqm4 running on STM32F429 M4-cortex CPU.
KW - Convolutional neural network
KW - Kyber
KW - Lattice-Based cryptography
KW - Plaintext-checking oracle
KW - Side-channel analysis
UR - http://www.scopus.com/inward/record.url?scp=85189649065&partnerID=8YFLogxK
U2 - 10.1007/978-981-97-1235-9_12
DO - 10.1007/978-981-97-1235-9_12
M3 - Conference contribution
AN - SCOPUS:85189649065
SN - 9789819712342
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 221
EP - 239
BT - Information Security and Cryptology – ICISC 2023 - 26th International Conference on Information Security and Cryptology, ICISC 2023, Revised Selected Papers
A2 - Seo, Hwajeong
A2 - Kim, Suhri
PB - Springer Science and Business Media Deutschland GmbH
T2 - 26th International Conference on Information Security and Cryptology on Information Security and Cryptology, ICISC 2023
Y2 - 29 November 2023 through 1 December 2023
ER -