Detecção de Botnets em Dispositivos IoTs baseado em LSTM Autoencoder

• Caio Maciel UFMG
• Anderson B. de Neira UFPR
• Ligia F. Borges UFMG
• Michele Nogueira UFMG / UFPR

DOI: https://doi.org/10.5753/sbseg.2023.233496

Abstract

In the face of the increasing adoption of Internet of Things (IoT), malicious actors exploit system loopholes to create software-infected networks (botnets). Botnets generate a series of service threats to security. Existing solutions specialize in attack scenarios. This decreases effectiveness in non-attack scenarios, increasing the number of incorrect detections, and generates more computational costs and rework. To deal with this problem, this work employs the LSTM Autoencoder combined with the majority vote technique. The proposal improves performance in scenarios with and without botnet attacks, reducing the number of false positives and avoiding costs. Preliminary results indicate an accuracy of 99.42% in detecting botnets, surpassing the current literature.

References

Baruah, S., Borah, D. J., and Deka, V. (2023). Detection of peer-to-peer botnet using machine learning techniques and ensemble learning algorithm. IJISP, 17(1):1–16.

Cunha, A. A., Borges, J. B., and Loureiro, A. A. (2022). Detecção de ataques de botnets em IoT via variational autoencoder. In CoUrb, pages 238–251. SBC.

e Silva, G. M., Neira, A., and Nogueira, M. (2022). Aprendizado profundo para a predição de ataques de negação de serviço distribuído. In SBRC, pages 475–488, Brasil. SBC.

Hochreiter, S. and Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8):1735–1780.

Jyoti, N. and Behal, S. (2021). A meta-evaluation of machine learning techniques for detection of ddos attacks. In INDIACom, pages 522–526. IEEE.

Meidan, Y., Bohadana, M., Mathov, Y., Mirsky, Y., Shabtai, A., Breitenbacher, D., and Elovici, Y. (2018). N-baiot—network-based detection of IoT botnet attacks using deep autoencoders. IEEE Pervasive Computing, 17(3):12–22.

Mirsky, Y., Doitshman, T., Elovici, Y., and Shabtai, A. (2018). Kitsune: an ensemble of autoencoders for online network intrusion detection. arXiv preprint arXiv:1802.09089.

Wang, C., Tindemans, S., Pan, K., and Palensky, P. (2020). Detection of false data injection attacks using the autoencoder approach. In PMAPS, pages 1–6. IEEE.