Side Channel Attack Fused with CNN_LSTM

doi:10.3778/j.issn.1002-8331.2111-0511

Abstract

Abstract: Based on the application of deep learning in side-channel attacks, the AES algorithm is first implemented in the Chipwhisperer platform, the corresponding energy trace is measured during its encryption process, and then the CPA technology is used to analyze the location of the point of interest, and model training is made for the point of interest. On the three network models convolutional neural network（CNN）, long-short-term memory network（LSTM） and CNN_LSTM hybrid model, combined with data preprocessing technology to train synchronous and asynchronous energy traces. The experimental results show that the accuracy of the three models in the synchronous state is equivalent. In addition, when the asynchronous data is gradually increased while ensuring the model training parameters remain unchanged, the accuracy of the training set and test set of the three models is decreasing, but the declining speed of the new hybrid model is the slowest. When the experimental asynchronous number is increased to 50, the accuracy can still be guaranteed to be above 90%, that is, the correct key can be recovered with almost one energy trace. Therefore, the CNN_LSTM model can better adapt to the situation where the energy trace occurs asynchronously.

Key words: advanced encryption standard（AES）, side channel attack, convolutional neural network, long short-term memory network, deep learning

摘要： 基于深度学习在侧信道攻击中的应用，在Chipwhisperer平台中实现AES算法，在其加密过程中测量相应能量迹，再利用CPA技术分析得出兴趣点位置，并针对兴趣点做出模型训练。在卷积神经网络（CNN），长短时记忆网络（LSTM）和CNN_LSTM混合模型三种网络模型上，结合数据预处理技术训练同步和异步能量迹。实验结果表明三种模型同步状态下的准确率相当，另外在保证模型训练参数不变的情况下逐渐增大异步数据时，三个模型训练集和测试集的准确率都在减少，但新提出的混合模型下降速度变化是最慢的，在实验异步数加大到50时，仍可以保证准确率在90%之上，即几乎一条能量迹就可恢复出正确密钥。所以，CNN_LSTM模型可以更好地适应能量迹发生异步的情况。

关键词: 高级加密标准（AES）, 侧信道攻击, 卷积神经网络, 长短时网络, 深度学习

PENG Pei, ZHANG Meiling, ZHENG Dong. Side Channel Attack Fused with CNN_LSTM[J]. Computer Engineering and Applications, 2023, 59(6): 268-276.

彭佩, 张美玲, 郑东. 融合CNN_LSTM的侧信道攻击[J]. 计算机工程与应用, 2023, 59(6): 268-276.

References

[1] QUISQUATER J J，SAMYDE D.Electromagnetic analysis （EMA）：measures and counter-measures for smart cards[C]//International Conference on Research in Smart Cards.Berlin，Heidelberg：Springer，2001：200-210.
[2] KOCHER P，JAFFE J，JUN B.Differential power analysis[C]//Annual International Cryptology Conference.Berlin，Heidelberg：Springer，1999：388-397.
[3] GOODFELLOW I，BENGIO Y，COURVILLE A.Deep learning[M].Cambridge：MIT press，2016.
[4] LECUN Y，BENGIO Y，HINTON G.Deep learning[J].Nature，2015，521（7553）：436-444.
[5] MARTINASEK Z，ZEMAN V.Innovative method of the power analysis[J].Journal of Adioengineering，2013，22（2）：586-594.
[6] MAGHREBI H，PORTIGLIATTI T，PROUFF E.Breaking cryptographic implementations using deep learning techniques[C]//International Conference on Security，Privacy，and Applied Cryptography Engineering.Heidelberg：Springer，2016：1-25.
[7] CAGLI E，DUMAS C，PROUFF E.Convolutional neural networks with data augmentation against jitter-based countermeasures[C]//Internationa Conference on Cryptographic Hardware and Embedded Systems，2017：45-68.
[8] BENADJILA R，PROUFF E，STRULLU R，et al.Study of deep learning techniques for side-channel analysis and introduction to ASCAD database[J].IACR Cryptology Eprint Archive，2018（53）：1-46.
[9] SAMIOTIS I P.Side-channel attacks using convolutional neural networks：a study on the performance of convolutional neural networks on side-channel data[D].Delft University of Technology，2018.
[10] TIMON B.Non-profiled deep learning-based side-channel attacks[J].IACR Transacyions on Cryptographic Hardqare and Embedded Systems，2019：107-131.
[11] ZAID G，BOSSUET L，HABRARD A，et al.Methodology for efficient CNN architectures in profiling attacks[J].Transactions on Cryptographic Hardware and Embedded Systems，2020（1）：1-36.
[12] WOUTERS L，ARRIBAS V，GIERLICHS B，et al.Revisiting a methodology for efficient CNNarchitectures in profiling attacks[J].Transactions on Cryptographic Hardware and Embedded Systems，2020（3）：147-168.
[13] 王俊年，朱斌，于文新，等.基于深度学习LSTM的侧信道分析[J].计算机工程，2021，47（10）：140-146.
WANG J N，ZHU B，YU W X，et al.Side channel analysis based on deep learning LSTM[J].Computer Engineering，2021，47（10）：140-146.
[14] WANG H，BRISFORS M，FORSMARK S，et al.How diversity affects deep-learning side-channel attacks[C]//2019 IEEE Nordic Circuits and Systems Conference（NORCAS）：NORCHIP and International Symposium of System-on-Chip（SoC），2019：1-7.
[15] 严春满，王铖.卷积神经网络模型发展及应用[J].计算机科学与探索，2021，15（1）：27-46.
YAN C M，WANG C.Development and application of convolutional neural network model[J].Journal of Frontiers of Computer Science and Technology，2021，15（1）：27-46.
[16] 陈平，汪平，胡红钢，等.基于SincNet的侧信道攻击[J].密码学报，2020，7（5）：583-594.
CHEN P，WANG P，HU H G，et al.SincNet-based side channel attack[J].Journal of Cryptography Research，2020，7（5）：583-594.
[17] HOWARD J，RUDER S.Universal language model fine-tuning for text classification[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics，2018：328-339.
[18] 兰朝凤，刘岩，赵宏运，等.基于波束形成的长短时记忆网络语音分离算法研究[J].电子与信息学报，2022，44（7）：2531-2538.
LAN C F，LIU Y，ZHAO H Y，et al.Research on longshort-term memory networks speech separation algorithm based on beamforming[J].Journal of Electronics and Information Technology，2022，44（7）：2531-2538.
[19] AKYLAS S，ATHANASIOS B，VASILIKI V，et al.Short-term net load forecasting with singular spectrum analysis and LSTM neural networks[J].Energies，2021，14（14）.
[20] DMITRI A，GAVRILOV M，JORDACHE A，et al.Preventing model overfitting and underfitting in convolutional neural networks[J].International Journal of Software Science and Computational Intelligence，2018，10（4）.
[21] ZHONG C W.Prediction of chineseport cargo volume based on BP algorithm[J].Journal of Physics：Conference Series，2021，1982（1）.
[22] LERMAN L，BONTEMPI G，MARKOWITCH O.A machine learning approach against a masked AES[J].Journal of Cryptographic Engineering，2015（5）：123-139.