Adversarial attacks and defenses for digital communication signals identification

Abstract

As modern communication technology advances apace, the digital communication signals identification plays an important role in cognitive radio networks, the communication monitoring and management systems. AI has become a promising solution to this problem due to its powerful modeling capability, which has become a consensus in academia and industry. However, because of the data-dependence and inexplicability of AI models and the openness of electromagnetic space, the physical layer digital communication signals identification model is threatened by adversarial attacks. Adversarial examples pose a common threat to AI models, where well-designed and slight perturbations added to input data can cause wrong results. Therefore, the security of AI models for the digital communication signals identification is the premise of its efficient and credible applications. In this paper, we first launch adversarial attacks on the end-to-end AI model for automatic modulation classification, and then we explain and present three defense mechanisms based on the adversarial principle. Next we present more detailed adversarial indicators to evaluate attack and defense behavior. Finally, a demonstration verification system is developed to show that the adversarial attack is a real threat to the digital communication signals identification model, which should be paid more attention in future research.

Keywords

Digital communication signals identification / AI model / Adversarial attacks / Adversarial defenses / Adversarial indicators

References

[1] A. Zhu, M. Ma, S. Guo, S. Yu, L. Yi, Adaptive multi-access algorithm for multi-service edge users in 5g ultra-dense heterogeneous networks, IEEE Trans. Veh. Technol. 70 (3) (2021) 2807-2821.

[2] Y. I. A. Al-Yasir, A. M. Abdulkhaleq, N. O. Parchin, I. T. Elfergani, J. Rodriguez, J. M. Noras, R. A. Abd-Alhameed, A. Rayit, R. Qahwaji, GReen and highly efficient MIMO transceiver system for 5g heterogenous networks, IEEE Trans. Green Commun. Netw. To appear. 6(1)(2022)500-511.

[3] Y. Siriwardhana, P. Porambage, M. Liyanage, M. Ylianttila, A survey on mobile augmented reality with 5G mobile edge computing: architectures, applications, and technical aspects, IEEE Commun. Surv. Tutor. 23 (2) (2021) 1160-1192.

[4] C. Zhang, Y.-L. Ueng, C. Studer, A. Burg, Artificial intelligence for 5g and beyond 5G: implementations, algorithms, and optimizations, IEEE J. Emerg. Select. Top. Circ. Syst. 10 (2) (2020) 149-163.

[5] M. Giordani, M. Polese, M. Mezzavilla, S. Rangan, M. Zorzi, Toward 6g networks: use cases and technologies, IEEE Commun. Mag. 58 (3) (2020) 55-61.

[6] N.A. Mousa, S.B. Sadkhan, Identif. digit. modul. signal cogn. radio netw. a surv.(2021) 311-314, https://doi.org/10.1109/BICITS51482.2021.9509927.

[7] S. Peng, S. Sun, Y.-D. Yao, A survey of modulation classification using deep learning: signal representation and data preprocessing, IEEE Transact. Neural Networks Learn. Syst. 33 (12) (2022) 7020-7038.

[8] Z. Lv, A.K. Singh, J. Li, Deep learning for security problems in 5G heterogeneous networks, IEEE Netw. 35 (2) (2021) 67-73.

[9] V. Monga, Y. Li, Y.C. Eldar, Algorithm unrolling: interpretable, efficient deep learning for signal and image processing, IEEE Signal Process. Mag. 38 (2) (2021) 18-44.

[10] C. Luo, J. Ji, Q. Wang, X. Chen, P. Li, Channel state information prediction for 5G wireless communications: a deep learning approach, IEEE Trans. Netw. Sci. Eng. 7 (1) (2020) 227-236.

[11] M.-Y. Chen, M.-H. Fan, L.-X. Huang, Ai-based vehicular network toward 6g and iot: deep learning approaches, ACM Trans. Manage. Inf. Syst. 13 (1)(2022)1-12.

[12] M.E. Morocho Cayamcela, W. Lim, Artificial intelligence in 5G technology: a survey,in:Proc. 2018 International Conference on Information and Communication Technology Convergence, Jeju Island, Korea, 2018, pp. 860-865.

[13] T.J. O'Shea, J. Corgan, T.C. Clancy, Convolutional radio modulation recognition networks, in: C. Jayne, L. Iliadis (Engineering Applications of Neural Networks,Eds.), Cham, 2016, pp. 213-226.

[14] T.J. O'Shea, T. Roy, T.C. Clancy, Over-the-air deep learning based radio signal classification, IEEE Journal of Selected Topics in Signal Processing 12 (1) (2018) 168-179.

[15] Y. Tu, Y. Lin, H. Zha, J. Zhang, Y. Wang, G. Gui, S. Mao, Large-scale real-world radio signal recognition with deep learning, Chin. J. Aeronaut. doi:https://doi.org/10.1016/j.cja.2021.08.016. URL https://www.sciencedirect.com/science/article/pii/S1000936121002934.

[16] S. Peng, H. Jiang, H. Wang, H. Alwageed, Y. Zhou, M.M. Sebdani, Y.-D. Yao, Modulation classification based on signal constellation diagrams and deep learning, IEEE Transact. Neural Networks Learn. Syst. 30 (3) (2019) 718-727.

[17] Y. Lin, Y. Tu, Z. Dou, L. Chen, S. Mao, Contour stella image and deep learning for signal recognition in the physical layer, IEEE Trans. Cogn. Commun. Netw. 7 (1)(2021) 34-46.

[18] Z. Ke, H. Vikalo, Real-time radio technology and modulation classification via an LSTM auto-encoder, IEEE Trans. Wireless Commun. 21 (1) (2022) 370-382.

[19] M.M.T. Abdelreheem, M.O. Helmi,Digitalmmodulation classification through time and frequency domain features using neural networks, in:Proc. 2012 IX International Symposium on Telecommunications, No. Oct., Bosnia and Herzegovina, Sarajevo, 2012, pp. 1-5.

[20] P. Qi, X. Zhou, S. Zheng, Z. Li, Automatic modulation classification based on deep residual networks with multimodal information, IEEE Trans. Cogn. Commun. Netw. 7 (1) (2021) 21-33.

[21] Q. Zhou, R. Zhang, J. Mu, H. Zhang, F. Zhang, X. Jing, AMCRN: few-shot learning for automatic modulation classification, IEEE Commun. Lett. 26 (3) (2022) 542-546.

[22] Y. Dong, X. Jiang, H. Zhou, Y. Lin, Q. Shi, SR2CNN: zero-shot learning for signal recognition, IEEE Trans. Signal Process. 69 (2021) 2316-2329.

[23] M. Wang, Y. Lin, Q. Tian, G. Si, Transfer learning promotes 6G wireless communications: recent advances and future challenges, IEEE Trans. Reliab. 70 (2)(2021) 790-807.

[24] Y. Lin, Y. Tu, Z. Dou, An improved neural network pruning technology for automatic modulation classification in edge devices, IEEE Trans. Veh. Technol. 69 (5) (2020) 5703-5706.

[25] S. Zhang, Y. Lin, Y. Tu, S. Mao, Electromagnetic signal modulation recognition technology based on lightweight deep neural network, J. Commun. 41 (11) (2020) 12.

[26] Y. Hu, W. Kuang, Z. Qin, K. Li, J. Zhang, Y. Gao, W. Li, K. Li, ARtificial intelligence security: threats and countermeasures, ACM Comput. Surv. 55 (1).

[27] Y. Yao, H. Li, H. Zheng, B.Y. Zhao, Latent backdoor attacks on deep neural networks, in: Proc. 2019 ACM SIGSAC Conference on Computer and Communications Security, London, UK, 2019, pp. 2041-2055.

[28] C. Szegedy, W. Zaremba, I. Sutskever, J. Bruna, D. Erhan, I. Goodfellow, R. Fergus, Intriguing Properties of Neural Networks, arXiv E-Prints, 2013 arXiv: 1312.6199arXiv:1312.6199.

[29] Y.E. Sagduyu, T. Erpek, Y. Shi,Adversarial Machine Learning for 5G Communications Security, arXiv E-Prints, 2021 arXiv: 2101.02656arXiv: 2101.02656.

[30] Z. Bao, Y. Lin, S. Zhang, Z. Li, S. Mao, Threat of adversarial attacks on dl-based iot device identification, IEEE Internet Things J. 9 (11) (2022) 9012-9024.

[31] M. Sadeghi, E.G. Larsson, Adversarial attacks on deep-learning based radio signal classification, IEEE Wireless Commun. Lett. 8 (1) (2019) 213-216.

[32] Y. Lin, H. Zhao, X. Ma, Y. Tu, M. Wang, Adversarial attacks in modulation recognition with convolutional neural networks, IEEE Trans. Reliab. 70 (1) (2021) 389-401.

[33] M. Usama, M. Asim, J. Qadir, A. Al-Fuqaha, M.A. Imran,Adversarial machine learning attack on modulation classification, in:Proc. 2019 UK, China Emerging Technologies, Glasgow, UK, 2019, pp. 1-4.

[34] B. Kim, Y.E. Sagduyu, K. Davaslioglu, T. Erpek, S. Ulukus, Channel-aware adversarial attacks against deep learning-based wireless signal classifiers, IEEE Trans. Wireless Commun. 21 (6) (2022) 3868-3880.

[35] B. Kim, Y.E. Sagduyu, K. Davaslioglu, T. Erpek, S. Ulukus, How to make 5G communications ”invisible”: adversarial machine learning for wireless privacy,in:Proc. 2020 54th Asilomar Conference on Signals, Systems, and Computers, Virtual Conference, 2020, pp. 763-767.

[36] D. Adesina, C.-C. Hsieh, Y.E. Sagduyu, L. Qian, Adversarial Machine Learning in Wireless Communications Using RF Data: A Review, arXiv E-Prints, 2020 arXiv: 2012.14392.

[37] S. Kokalj-Filipovic, R. Miller, G. Vanhoy, Adversarial examples in rf deep learning: detection and physical robustness, in: PRoc. IEEE GlobalSIP’19, Canada, Ottawa, 2019, pp. 1-5.

[38] Z. Yuan, J. Zhang, Y. Jia, C. Tan, T. Xue, S. Shan, Meta Gradient Adversarial Attack, 2021, pp. 7748-7757.

[39] R. Sahay, D.J. Love, C.G. Brinton, Robust Automatic Modulation Classification in the Presence of Adversarial Attacks, 2021, pp. 22955-22967.

[40] Y. Lin, H. Zhao, Y. Tu, S. Mao, Z. Dou, Threats of adversarial attacks in DNN-based modulation recognition, in: Proc. IEEE INFOCOM, vol. 2020, Virtual Conference, 2020, pp. 2469-2478.

[41] I.J. Goodfellow, J. Shlens, C. Szegedy, Explaining and Harnessing Adversarial Examples, arXiv E-Prints, 2014 arXiv:1412.6572.

[42] A. Madry, A. Makelov, L. Schmidt, D. Tsipras, A. Vladu, Towards Deep Learning Models Resistant to Adversarial Attacks, arXiv E-Prints, 2017 arXiv:1706.06083.

[43] L. Schmidt, S. Santurkar, D. Tsipras, K. Talwar, A. Madry, Adversarially Robust Generalization Requires More Data, arXiv E-Prints, 2018 arXiv: 1804.11285.

[44] H. Qin, R. Gong, X. Liu, X. Bai, J. Song, N. Sebe, Binary neural networks: a survey, Pattern Recogn. 105 (2020) 107281.

[45] S. Zhang, Y. Lin, Z. Bao, J. Fu, A Lightweight Modulation Classification Network Resisting White Box Gradient Attacks, Hindawi Security and Communication Networks, 2021.

[46] A. Galloway, G.W. Taylor, M. Moussa, Attacking Binarized Neural Networks, arXiv E-Prints, 2017 arXiv:1711.00449.