DL-SAFE: Proteção Baseada em Aprendizado Profundo para Detecção de Botnets na Borda

• Lucas Chagas de Brito Guimarães UFRJ
• Rodrigo de Souza Couto UFRJ

DOI: https://doi.org/10.5753/sbseg_estendido.2023.235788

Abstract

IoT (Internet of Things) devices are fundamental to multiple sectors, such as smart homes, cities, and grids. However, the existence of billions of devices with limited computing power makes them ideal targets for botnets. This paper proposes DL-SAFE, a tool for real-time traffic classification in edge environments using Open Argus and Pytorch. The tool also allows the evaluation of neural network architectures using 3 types of layers. The results demonstrate the tools’ effectiveness, obtaining precision and recall values greater than 99% for multiple models. With the throughput tests it can be seen that performance varies greatly according to the architecture, with half the evaluated models processing more than 1000 network flows per second.

References

Alicia Hope (2021). Russian Internet Giant Yandex Wards off the Largest Botnet DDoS Attack in History. available at: [link].

Alkadi, O., Moustafa, N., Turnbull, B., and Choo, K.-K. R. (2020). A deep blockchain framework-enabled collaborative intrusion detection for protecting iot and cloud networks. IEEE Internet of Things Journal, 8(12):9463–9472.

Catalin Cimpanu (2021). Microsoft said it mitigated a 2.4 Tbps DDoS attack. available at: [link].

Cisco (2018). Cisco Annual Internet Report (2018–2023). available at: [link].

Ferrag, M. A. and Maglaras, L. (2019). Deepcoin: A novel deep learning and blockchain-based energy exchange framework for smart grids. IEEE Transactions on Engineering Management, 67(4):1285–1297.

Ferrag, M. A., Maglaras, L., Moschoyiannis, S., and Janicke, H. (2020). Deep learning for cyber security intrusion detection: Approaches, datasets, and comparative study. Journal of Information Security and Applications, 50:102419.

Jan, S., Masoodi, F., and Bamhdi, A. (2022). Effective intrusion detection in iot environment: Deep learning approach. In SCRS Conference Proceedings on Intelligent Systems, pages 495–502.

Koroniotis, N., Moustafa, N., Sitnikova, E., and Turnbull, B. (2019). Towards the development of realistic botnet dataset in the internet of things for network forensic analytics: Bot-iot dataset. Future Generation Computer Systems, 100:779–796.

Mainuddin, M., Duan, Z., and Dong, Y. (2021). Network traffic characteristics of iot devices in smart homes. In International Conference on Computer Communications and Networks (ICCCN), pages 1–11.

Neshenko, N., Bou-Harb, E., Crichigno, J., Kaddoum, G., and Ghani, N. (2019). Demystifying iot security: An exhaustive survey on iot vulnerabilities and a first empirical look on internet-scale iot exploitations. IEEE Communications Surveys & Tutorials, 21(3):2702–2733.

Popoola, S. I., Adebisi, B., Ande, R., Hammoudeh, M., Anoh, K., and Atayero, A. A. (2021a). smote-drnn: A deep learning algorithm for botnet detection in the internet-of-things networks. Sensors, 21(9):2985.

Popoola, S. I., Ande, R., Adebisi, B., Gui, G., Hammoudeh, M., and Jogunola, O. (2021b). Federated deep learning for zero-day botnet attack detection in iot-edge devices. IEEE Internet of Things Journal, 9(5):3930–3944.

Saurabh, K., Sood, S., Kumar, P. A., Singh, U., Vyas, R., Vyas, O., and Khondoker, R. (2022). LBDMIDS: LSTM based deep learning model for intrusion detection systems for iot networks. In IEEE World AI IoT Congress (AIIoT), pages 753–759.

Shao, Z., Yuan, S., and Wang, Y. (2021). Adaptive online learning for iot botnet detection. Information Sciences, 574:84–95.

Velasco-Mata, J., González-Castro, V., Fidalgo, E., and Alegre, E. (2023). Real-time botnet detection on large network bandwidths using machine learning. Scientific Reports, 13(1):4282.