Evaluating Adversarial Attacks in Applications of Brain-Computer Interfaces

• Beatriz Costa Federal University of Rio Grande (FURG)
• André Riker Federal University of Pará (UFPA) https://orcid.org/0009-0006-8864-9948
• Bruno L. Dalmazo Federal University of Rio Grande (FURG) https://orcid.org/0000-0002-6996-7602

DOI: https://doi.org/10.5753/errc.2024.4583

Abstract

Brain-computer interfaces (BCI) are systems that capture brain signals through techniques such as electroencephalography (EEG), processing these signals for various applications, especially in the control of devices for people with motor limitations. Despite the benefits, there are security concerns, including adversary and cybersecurity attacks. Protecting users’ personal data is crucial, requiring continuous research into cybersecurity to mitigate risks and ensure privacy in BCI applications. During the evaluation, we can identify negative effects on data classification produced by adversarial attacks. Due to the emergence of brain-computer interaction devices on the market, which accesses users’ brain waves and analyzes them for various purposes, it is necessary to analyze the security of these devices. Therefore, this work aims to emulate and analyze adversarial attacks in classifiers of brain-computer interface devices

Keywords: Security, brain-computer interface, adversarial attacks

References

Bernal, S. L. et al. Security in Brain-Computer Interfaces: State-of-the-Art, Opportunities, and Future Challenges. ACM Comput. Surv., Association for Computing Machinery, New York, NY, USA, v. 54, n. 1, jan. 2021. ISSN 0360-0300. DOI: 10.1145/3427376.

Dalmazo, B. L.; Vilela, J. P.; Curado, M. Performance analysis of network traffic predictors in the cloud. Journal of Network and Systems Management, Springer, v. 25, p. 290–320, 2017.

Dalmazo, B. L.; Vilela, J. P.; Curado, M. Triple-Similarity Mechanism for alarm management in the cloud. Computers & Security, v. 78, p. 33–42, 2018. ISSN 0167-4048. DOI: 10.1016/j.cose.2018.05.016. Disponível em: [link].

Neurable. Neurable. 2024. Disponível em: [link].

Neuralink. Neuralink. 2024. Disponível em: [link].

Neurosity. Neurosity. 2024. Disponível em: [link].

Jiang, X.; Zhang, X.; Wu, D. Active Learning for Black-Box Adversarial Attacks in EEG-Based Brain-Computer Interfaces. In: 2019 IEEE Symposium Series on Computational Intelligence (SSCI). 2019. P. 361–368. DOI: 10.1109/SSCI44817.2019.9002719.

Jung, J. et al. Generative Perturbation Network for Universal Adversarial Attacks on Brain-Computer Interfaces. IEEE Journal of Biomedical and Health Informatics, v. 27, n. 11, p. 5622–5633, 2023. DOI: 10.1109/JBHI.2023.3303494.

Upadhayay, B.; Behzadan, V. Adversarial Stimuli: Attacking Brain-Computer Interfaces via Perturbed Sensory Events. In: 2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC). 2023. P. 3061–3066. DOI: 10.1109/SMC53992.2023.10394505.

Feng, B.; Wang, Y.; Ding, Y. Saga: Sparse Adversarial Attack on EEG-Based Brain Computer Interface. In: ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 2021. P. 975–979. DOI: 10.1109/ICASSP39728.2021.9413507.

Aissa, N. E. H. S. B. et al. Enhancing EEG Signal Classifier Robustness Against Adversarial Attacks Using a Generative Adversarial Network Approach. IEEE Internet of Things Magazine, v. 7, n. 3, p. 44–49, 2024. DOI: 10.1109/IOTM.001.2300262.

Li, Y. et al. Adversarial Training for the Adversarial Robustness of EEG-Based Brain-Computer Interfaces. In: 2022 IEEE 32nd International Workshop on Machine Learning for Signal Processing (MLSP). 2022. P. 1–6. DOI: 10.1109/MLSP55214.2022.9943479.

Aissa, N. E. H. S. B. et al. Robust Detection of Adversarial Attacks for EEG-based Motor Imagery Classification using Hierarchical Deep Learning. In: 2023 15th International Conference on Innovations in Information Technology (IIT). 2023. P. 156–161. DOI: 10.1109/IIT59782.2023.10366492.

Debie, E.; Moustafa, N.; Whitty, M. T. A Privacy-Preserving Generative Adversarial Network Method for Securing EEG Brain Signals. In: 2020 International Joint Conference on Neural Networks (IJCNN). 2020. P. 1–8. DOI: 10.1109/IJCNN48605.2020.9206683.

Leite, L. et al. Federated Learning under Attack: Improving Gradient Inversion for Batch of Images. In: ANAIS do XXIV Simpósio Brasileiro de Segurança da Informação e de Sistemas Computacionais. São José dos Campos/SP: SBC, 2024. P. 794–800. DOI: 10.5753/sbseg.2024.241680. Disponível em: [link].

Kaya, M. et al. A large electroencephalographic motor imagery dataset for electroencephalographic brain computer interfaces. 2018. Wiki do abnTeX2. Disponível em: [link].

Beltrán, E. T. M. et al. SecBrain: A Framework to Detect Cyberattacks Revealing Sensitive Data in Brain-Computer Interfaces. In: ADVANCES in Malware and Data-Driven Network Security. IGI Global, 2022. P. 176–198.

Martínez Beltrán, E. T. et al. Noise-based cyberattacks generating fake P300 waves in brain–computer interfaces. Cluster Computing, Springer, p. 1–16, 2021.

Amorim, M. et al. Systematic Review of Aggregation Functions Applied to Image Edge Detection. Axioms, v. 12, n. 4, 2023. ISSN 2075-1680. DOI: 10.3390/axioms12.040330. Disponível em: [link].

Lawhern, V. J. et al. EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces. Journal of Neural Engineering, IOP Publishing, v. 15, n. 5, p. 056013, jul. 2018. DOI: 10.1088/1741-2552/aace8c.