Minimum Power Adversarial Attacks in Communication Signal Modulation Classification with Deep Learning
Tóm tắt
Integrating cognitive radio (CR) technique with wireless networks is an effective way to solve the increasingly crowded spectrum. Automatic modulation classification (AMC) plays an important role in CR. AMC significantly improves the intelligence of CR system by classifying the modulation type and signal parameters of received communication signals. AMC can provide more information for decision making of the CR system. In addition, AMC can help the CR system dynamically adjust the modulation type and coding rate of the communication signal to adapt to different channel qualities, and the AMC technique help eliminate the cost of broadcast modulation type and coding rate. Deep learning (DL) has recently emerged as one most popular method in AMC of communication signals. Despite their success, DL models have recently been shown vulnerable to adversarial attacks in pattern recognition and computer vision. Namely, they can be easily deceived if a small and carefully designed perturbation called an adversarial attack is imposed on the input, typically an image in pattern recognition. Owing to the very different nature of communication signals, it is interesting yet crucially important to study if adversarial perturbation could also fool AMC. In this paper, we make a first attempt to investigate how we can design a special adversarial attack on AMC. we start from the assumption of a linear binary classifier which is further extended to multi-way classifier. We consider the minimum power consumption that is different from existing adversarial perturbation but more reasonable in the context of AMC. We then develop a novel adversarial perturbation generation method that leads to high attack success to communication signals. Experimental results on real data show that the method is able to successfully spoof the 11-class modulation classification at a model with a minimum cost of about − 21 dB in automatic modulation classification task. The visualization results demonstrate that the adversarial perturbation manifests in the time domain as imperceptible undulations of the signal, and in the frequency domain as small noise outside the signal band.
Tài liệu tham khảo
Jin X, Sun J, Zhang R, Zhang Y, Zhang C. Deep learning for an effective non-orthogonal multiple access scheme. IEEE Trans Mobile Comput. 2018;17(12):2925–38.
Khan AA, Rehmani MH, Reisslein M. Cognitive radio for smart grids: survey of architectures, spectrum sensing mechanisms, and networking protocols. IEEE Commun Surv Tut. 2015;18(1):860–98.
Ul Hassan M, Rehmani MH, Rehan M, et al. Differential privacy in cognitive radio networks: a comprehensive survey. Cogn Comput. 2022. https://doi.org/10.1007/s12559-021-09969-9.
Shi C, Dou Z, Lin Y, Li W. Dynamic threshold-setting for RF powered cognitive radio networks in non-Gaussian noise. Physical Communication. 2018;27(1):99–105.
Wang H, Li J, Guo L, Dou Z, Lin Y, Zhou R. Fractal complexity based feature extraction algorithm of communication signals. Fractals. 2017;25(4):1740008–20.
Zhang Z, Guo X, Lin Y. Trust management method of D2D communication based on RF fingerprint identification. IEEE Access. 2018;6:66082–7.
Wang Y, Liu M, Yang J, Gui G. Data-driven deep learning for automatic modulation recognition in cognitive radios. IEEE Trans Veh Technol. 2019;68(4):4074–7.
Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. In Proc IEEE Conf Comp Vision Pattern Recogn. 2015;1(1):3431–3440.
Meng F, Chen P, Wu L. Automatic modulation classification: a deep learning enabled approach. IEEE Trans Vehicular Technol. 2018.
Zhao Y, Wang X, Lin Z, et al. multi classifier fusion for open set specific emitter identification. Remote Sens. 2022;14(9):2226. https://doi.org/10.3390/rs14092226.
Sun L, Wang X, Huang Z, Li B. Radio frequency fingerprint extraction based on feature inhomogeneity. IEEE Internet Things J. 2022. https://doi.org/10.1109/JIOT.2022.3154595.
Sun L, Wang X, Zhao Y, Huang Z, Chun Du. Intrinsic low-dimensional nonlinear manifold structure of radio frequency signals. IEEE Commun Lett. 2022. https://doi.org/10.1109/LCOMM.2022.3173990.
Huang K, Hussain A, Wang QF, et al. Deep learning: fundamentals, theory and applications. Springer, ISBN 978–3–030–06072–5. 2019.
Szegedy C, et al. Intriguing properties of neural networks. In Proc Int Conf Learn Repr. 2015;1–10.
Ke D, et al. Application of adversarial examples in communication modulation classification. 2019 International Conference on Data Mining Workshops (ICDMW). 2019.
Goodfellow IJ, Shlens J, Szegedy C. Explaining and harnessing adversarial examples. Comp Sci. 2014.
Lyu C, Huang K, Liang HN. A Unified Gradient Regularization Family for Adversarial Examples. ICDM. 2015.
Papernot N, Mcdaniel P, Jha S, et al. The limitations of deep learning in adversarial settings. IEEE Eur symp sec privacy (EuroS&P). 2016.
Carlini N, Wagner D. Towards evaluating the robustness of neural networks. IEEE. 2017.
DeepSig, Deepsig dataset: Radioml 2016.10a, 2016. Available: https://www.deepsig.io/datasets.
Song L, Qian X, Li H, Chen Y. Pipelayer: a pipelined reram-based accelerator for deep learning. In Proc IEEE Int Symp High Perform Comput Arch. 2017;1(1):541–552.
Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. 2014. arXiv preprint arXiv:1409.1556.
He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In Proc IEEE Conf Comp Vis Pattern Recogn. 2016;770–778.
Zhang S, Huang K, Zhu J, Liu Y. Manifold adversarial training for supervised and semi-supervised learning. Neural Netw. 2021;140:282–93.
Madry A, et al. Towards deep learning models resistant to adversarial attacks. 2017. arXiv preprint arXiv:1706.06083.