Deep learning network attack trends using side channel analysis

부채널 분석을 이용한 딥러닝 네트워크 공격 동향

  • Duk-Young Kim (Dept. of IT Convergence Security, Han-sung University) ;
  • Hyun-Ji Kim (Dept. of Information Computer Engineering, Han-sung University) ;
  • Hyun-Jun Kim (Dept. of Information Computer Engineering, Han-sung University) ;
  • Hwa-Jeong Seo (Dept. of IT Convergence Security, Han-sung University)
  • 김덕영 (한성대학교 융합보안학과 ) ;
  • 김현지 (한성대학교 정보컴퓨터공학과 ) ;
  • 김현준 (한성대학교 정보컴퓨터공학과 ) ;
  • 서화정 (한성대학교 융합보안학과 )
  • Published : 2024.05.23

Abstract

최근 빠른 속도로 개발되고 있는 인공지능 기술은 여러 산업 분야에서 활용 되고 있다. 그러나 최근 딥러닝 네트워크에 대한 부채널 공격 기법들이 등장하고 있으며, 이는 해당 모델을 재구현하여 자율 주행 자동차에 대한 해킹 등과 같이 치명적인 보안 위협이 될 수 있으므로 이에 대한 이해와 대응책이 필요하다. 본 논문에서는 딥러닝 네트워크에 대한 부채널 공격 기법 동향에 대해 살펴보고, 이에 대한 대응 기술 또한 함께 알아본다.

Keywords

Acknowledgement

This work was supported by Institute for Information & communications Technology Promotion(IITP) grant funded by the Korea government(MSIT) (No.2018-0-00264, Research on Blockchain Security Technology for IoT Services, 50%) and this work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government(MSIT)(No.2022-0-00627, Development of Lightweight BIoT technology for Highly Constrained Devices, 50%).

References

  1. LeCun, Yann, Yoshua Bengio, and Geoffrey Hinton. "Deep learning." nature 521.7553 (2015): 436-444. 
  2. Joye, Marc, and Francis Olivier. "Side-Channel Analysis." (2011):1198-1204. 
  3. Mangard, Stefan, Elisabeth Oswald, and Thomas Popp. Power analysis attacks: Revealing the secrets of smart cards. Vol. 31. Springer Science & Business Media, 2008. 
  4. Kocher, Paul, Joshua Jaffe, and Benjamin Jun. "Differential power analysis." Advances in Cryptology-CRYPTO'99: 19th Annual International Cryptology Conference Santa Barbara, California, USA, August 15-19, 1999 Proceedings 19. Springer Berlin Heidelberg, 1999. 
  5. Brier, Eric, Christophe Clavier, and Francis Olivier. "Correlation power analysis with a leakage model." Cryptographic Hardware and Embedded Systems-CHES 2004: 6th International Workshop Cambridge, MA, USA, August 11-13, 2004. Proceedings 6. Springer Berlin Heidelberg, 2004. 
  6. Yoshida, Kota, et al. "Model reverse-engineering attack using correlation power analysis against systolic array based neural network accelerator." 2020 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE, 2020. 
  7. Yoshida, Kota, et al. "Model reverse-engineering attack against systolic-array-based dnn accelerator using correlation power analysis." IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences 104.1 (2021): 152-161. 
  8. Maji, Saurav, Utsav Banerjee, and Anantha P. Chandrakasan. "Leaky nets: Recovering embedded neural network models and inputs through simple power and timing side-channels-Attacks and defenses." IEEE Internet of Things Journal 8.15 (2021): 12079-12092. 
  9. Liu, Yuntao, Dana Dachman-Soled, and Ankur Srivastava. "Mitigating reverse engineering attacks on deep neural networks." 2019 IEEE Computer Society Annual Symposium on VLSI (ISVLSI). IEEE, 2019. 
  10. Athanasiou, Konstantinos, et al. "Masking feedforward neural networks against power analysis attacks." proceedings on privacy enhancing technologies (2022).