Abstract

Cryptanalytic side-channel analysis exposes potential vulnerabilities in cryptographic systems by exploiting physical leakages such as power consumption or electromagnetic emissions during cryptographic operations. These side-channel attacks (SCA) aim to recover sensitive information, including encryption keys, by analyzing subtle variations in power traces or radiated signals. In recent years, deep learning (DL) techniques have demonstrated exceptional performance in assessing embedded system security due to their ability to model complex patterns with high accuracy. However, challenges such as high computational costs and limited prediction accuracy still persist in SCA contexts. This study introduces a novel approach named grasshopper search rescue optimization (GSRO), based on grasshopper optimization algorithm (GOA) and adapted for search and rescue scenarios, to optimize the hyperparameters of a multi-layer perceptron (MLP) classifier. The GSRO-enhanced MLP model correlates power traces with the probability of intermediate cryptographic values, facilitating more accurate and efficient SCA detection. Furthermore, power traces are converted into intermediate representations to enhance learning effectiveness. The proposed method achieved an accuracy of 92%, outperforming existing models with reported accuracies between 85% and 89%. It also demonstrated a reduced error rate of 5%, compared to 8%–10% in conventional models. Moreover, the model achieved a Kling-Gupta Efficiency (KGE) score of 0.79, indicating superior capability in capturing both variability and agreement, compared to the average KGE score of 0.75 in state-of-the-art methods. These results confirm that the GSRO based optimization framework significantly improves the performance and robustness of deep learning models in side-channel analysis.

Keywords

Side-channel analysis (SCA), Grasshopper optimization algorithm (GOA), Hyperparameter optimization, Multi-layer perceptron (MLP), Deep learning, Cryptographic key recovery.

References

[1] Tehranipoor M, Nalla Anandakumar N, Farahmandi F. Power analysis attacks on AES. In hardware security training, hands-on! 2023 (pp.137-61). Cham: Springer International Publishing.

[2] Masure L, Strullu R. Side-channel analysis against ANSSI’s protected AES implementation on ARM: end-to-end attacks with multi-task learning. Journal of Cryptographic Engineering. 2023; 13(2):129-47.

[3] Imafuku K, Kawamura S, Nozaki H, Sakamoto J, Osuka S. Non-profiled deep learning-based side-channel analysis with only one network training. IEEE Access. 2023; 11:83221-31.

[4] Jovic A, Jap D, Papachristodoulou L, Heuser A. Traditional machine learning methods for side-channel analysis. In security and artificial intelligence: a crossdisciplinary approach 2022 (pp. 25-47). Cham: Springer International Publishing.

[5] Bae D, Hwang J, Ha J. Deep learning-based attacks on masked AES implementation. Journal of Internet Technology. 2022; 23(4):897-902.

[6] Staib M, Moradi A. Deep learning side-channel collision attack. IACR Transactions on Cryptographic Hardware and Embedded Systems. 2023; 2023(3):422-44.

[7] Barenghi A, Carrera D, Mella S, Pace A, Pelosi G, Susella R. Profiled side channel attacks against the RSA cryptosystem using neural networks. Journal of Information Security and Applications. 2022; 66:103122.

[8] Do NT, Hoang VP, Doan VS, Pham CK. On the performance of non‐profiled side channel attacks based on deep learning techniques. IET Information Security. 2023; 17(3):377-93.

[9] Afzal S, Ziapour BM, Shokri A, Shakibi H, Sobhani B. Building energy consumption prediction using multilayer perceptron neural network-assisted models; comparison of different optimization algorithms. Energy. 2023; 282:128446.

[10] Bhure C, Nicholas GS, Saqib F. Machine learning applications and attacks using side channel analysis. In AI-enabled electronic circuit and system design: from ideation to utilization 2024 (pp. 387-426). Cham: Springer Nature Switzerland.

[11] Rijsdijk J, Wu L, Perin G, Picek S. Reinforcement learning for hyperparameter tuning in deep learning-based side-channel analysis. IACR Transactions on Cryptographic Hardware and Embedded Systems. 2021: 677-707.

[12] Wu L, Perin G, Picek S. I choose you: automated hyperparameter tuning for deep learning-based side-channel analysis. IEEE Transactions on Emerging Topics in Computing. 2022; 12(2):546-57.

[13] Ou Y, Li L. Side-channel analysis attacks based on deep learning network. Frontiers of Computer Science. 2022; 16(2):162303.

[14] Ghandali S, Ghandali S, Tehranipoor S. Deep K-TSVM: a novel profiled power side-channel attack on AES-128. IEEE Access. 2021; 9:136448-58.

[15] Hu F, Wang H, Wang J. Multi-leak deep-learning side-channel analysis. IEEE Access. 2022; 10:22610-21.

[16] Picek S, Perin G, Mariot L, Wu L, Batina L. Sok: deep learning-based physical side-channel analysis. ACM Computing Surveys. 2023; 55(11):1-35.

[17] Park D, Choi M, Kim G, Bae D, Kim H, Hong S. Stealing keys from hardware wallets: a single trace side-channel attack on elliptic curve scalar multiplication without profiling. IEEE Access. 2023; 11:44578-89.

[18] Li D, Li L, Ou Y. Side-channel analysis based on Siamese neural network. The Journal of Supercomputing. 2024; 80(4):4423-50.

[19] He P, Zhang Y, Gan H, Ma J, Zhang H. Side-channel attacks based on attention mechanism and multi-scale convolutional neural network. Computers and Electrical Engineering. 2024; 119:109515.

[20] Li Y, Huang Y, Jia F, Zhao Q, Tang M, Ren S. A gradient deconvolutional network for side-channel attacks. Computers & Electrical Engineering. 2022; 98:107686.

[21] Chang L, Wei Y, He S, Pan X. Research on side-channel analysis based on deep learning with different sample data. Applied Sciences. 2022; 12(16):1-23.

[22] Li L, Ou Y. A deep learning-based side channel attack model for different block ciphers. Journal of Computational Science. 2023; 72:102078.

[23] Ahmed AA, Hasan MK, Alqahtani A, Islam S, Pandey B, Rzayeva L, et al. Deep learning based side-channel attack detection for mobile devices security in 5G networks. Tsinghua Science and Technology. 2024; 30(3):1012-26.

[24] Hameed F, Alkhzaimi H. Advanced side-channel profiling attacks with deep neural networks: a hill climbing approach. Electronics. 2024; 13(17):1-22.

[25] Ahmed AA, Hasan MK, Memon I, Aman AH, Islam S, Gadekallu TR, et al. Secure AI for 6G mobile devices: deep learning optimization against side-channel attacks. IEEE Transactions on Consumer Electronics. 2024; 70(1):3951-9.

[26] Hameed F, Ramesh SM, Alkhzaimi H. Improved hybrid bagging resampling framework for deep learning-based side-channel analysis. Computers. 2024; 13(8):1-29.

[27] Deng T, Wang H, He D, Xiong N, Liang W, Wang J. Multi-dimensional fusion deep learning for side channel analysis. Electronics. 2023; 12(23):1-16.

[28] Liu W, Zhang Y, Tang Y, Wang H, Wei Q. Alsca: a framework for using auxiliary learning side-channel attacks to model pufs. IEEE Transactions on Information Forensics and Security. 2022; 18:804-17.

[29] Negabi I, Ait EAS, El AS, Raissouni N. Deep learning-based power analysis attack for extracting AES keys on ATmega328P microcontroller. Arabian Journal for Science and Engineering. 2024; 49(3):4197-208.

[30] Kwon D, Hong S, Kim H. Optimizing implementations of non-profiled deep learning-based side-channel attacks. IEEE Access. 2022; 10:5957-67.

[31] Li Y, Huang Y, Guo Y, Ge C, Kong F, Ren Y. Adaptive chosen-plaintext deep learning-based side-channel analysis. IEEE Internet of Things Journal. 2024; 12(1):174-85.

[32] Hoang VP, Do NT, Doan VS. Performance analysis of deep learning based non-profiled side channel attacks using significant hamming weight labeling. Mobile Networks and Applications. 2023; 28(3):1187-96.

[33] Zunaidi MR, Sayakkara A, Scanlon M. Revealing IoT cryptographic settings through electromagnetic side-channel analysis. Electronics. 2024; 13(8):1-22.

[34] Subbareddy R, Selvan PT. Enhancing data security in cloud computing: a blockchain-based feistel cipher encryption and multiclass vector side-channel attack detection approach. International Journal of Advanced Technology and Engineering Exploration. 2024; 11(112):354-72.

[35] Wu L, Rezaeezade A, Ali-pour A, Perin G, Picek S. Leakage model-flexible deep learning-based side-channel analysis. IACR Communications in Cryptology. 2024; 1(3):1-26.

[36] Do NT, Hoang VP, Doan VS. A novel non-profiled side channel attack based on multi-output regression neural network. Journal of Cryptographic Engineering. 2024; 14(3):427-39.

[37] Krček M, Perin G. Autoencoder-enabled model portability for reducing hyperparameter tuning efforts in side-channel analysis. Journal of Cryptographic Engineering. 2024; 14(3):475-97.

[38] Abbas AA, Kamrul HM, Azman MNS, Hafizah AA. Design of time-delay convolutional neural networks (TDCNN) model for feature extraction for side-channel attacks. International Journal of Computing and Digital Systems. 2024; 16(1):341-51.

[39] Wu L, Won YS, Jap D, Perin G, Bhasin S, Picek S. Ablation analysis for multi-device deep learning-based physical side-channel analysis. IEEE Transactions on Dependable and Secure Computing. 2023; 21(3):1331-41.

[40] Muhammad AA, Alzuabidi IA, Ahmed AA, Abdulkadir RA. Adaptive optimization of deep learning models on AES based large side channel attack data. Alkadhim Journal for Computer Science. 2024; 2(1):72-85.

[41] Bae D, Park D, Kim G, Choi M, Lee N, Kim H, et al. Autoscaled-wavelet convolutional layer for deep learning-based side-channel analysis. IEEE Access. 2023; 11:95381-95.

[42] Liu A, Wang A, Sun S, Wei C, Ding Y, Wang Y, Zhu L. CL-SCA: leveraging contrastive learning for profiled side-channel analysis. IEEE Transactions on Information Forensics and Security. 2024; 14(8):1-14.

[43] Meraihi Y, Gabis AB, Mirjalili S, Ramdane-cherif A. Grasshopper optimization algorithm: theory, variants, and applications. IEEE Access. 2021; 9:50001-24.

[44] Wu L, Wu J, Wang T. Enhancing grasshopper optimization algorithm (GOA) with levy flight for engineering applications. Scientific Reports. 2023; 13(1):1-59.

[45] Wang C, Li J, Rao H, Chen A, Jiao J, Zou N, et al. Multi-objective grasshopper optimization algorithm based on multi-group and co-evolution. Mathematical Biosciences and Engineering. 2021; 18(3):2527-61.

Cite this article