Research on IoT Intrusion Detection Method Based on SSA and Transformer-BiGRU

Cui Zhongyuan(Zhengzhou University, Zhengzhou City)
Meng Dechao(Henan Province Sports Lottery Administration Center)

Abstract

The rapid advancement of Internet of Things (IoT) technology has enabled widespread deployment of IoT devices in critical domains including industrial control, smart cities, and intelligent transportation. Nevertheless, device heterogeneity and system openness create significant vulnerabilities to malicious attacks. Additional challenges such as intrusion traffic dynamics and class imbalance further degrade conventional detection models. This paper proposes an intrusion detection method based on collaborative optimization integrating Singular Spectrum Analysis (SSA) with a hybrid Transformer-BiGRU architecture. Our approach employs SSA with an enhanced fitness function for adaptive feature selection. A hybrid Transformer-BiGRU model is then constructed with SSA-optimized hyperparameters. Experimental evaluation using the CIC-IDS-2017 and CIC-IoT2023 datasets demonstrates the model's effectiveness, achieving classification accuracies of 96.72% and 97.83% respectively. These results confirm superior performance compared to existing approaches.

 

Keywords

Internet of Things Security; Intrusion Detection; Singular Spectrum Analysis (SSA); Transformer; Feature Optimization

References

Khraisat A, Gondal I, Vamplew P, et al. Survey of intrusion detection systems: techniques, datasets and challenges[J]. Cybersecurity, 2019, 2(1): 20.

C. Yin, Y. Zhu, J. Fei, and X. He, “A deep learning approach for intrusion detection using recurrent neural networks,” IEEE Access, vol. 5, pp. 21954–21961, 2017.

X. Wang et al., “A hybrid intrusion detection system based on Transformer and BiLSTM,” Future Generation Computer Systems, vol. 125, pp. 238–250, Dec. 2021.

A. Vaswani et al., “Attention is all you need,” in Advances in Neural Information Processing Systems, 2017, pp. 5998–6008.

Y. Zhang, Q. Wang, and L. Shen, “Network traffic analysis for industrial IoT anomaly detection using sparrow search algorithm and deep learning,” IEEE Internet of Things Journal, vol. 9, no. 18, pp. 18005–18015, Sep. 2022.

Y. Zhang, L. Shen, and H. Li, “A hybrid KF-SSA model for efficient intrusion detection in IoT networks,” Computer Networks, vol. 215, p. 109231, Oct. 2022.

N. Ashraf, S. R. Islam, and M. S. Hossain, “Global context-aware intrusion detection in IoT networks using self-attention mechanisms,” Computer Communications, vol. 190, pp. 176–186, May 2022.

H. Wang, Z. Liu, and Y. Zhang, “A lightweight multi-scale Transformer for intrusion detection using discrete wavelet transform in edge computing,” IEEE Transactions on Industrial Informatics, vol. 19, no. 2, pp. 1678–1687, Feb. 2023.

F. Li, M. Xu, and P. Wang, “A Transformer-BiGRU model for efficient and accurate network intrusion detection,” Journal of Network and Computer Applications, vol. 210, p. 103540, Jan. 2023.

X. Zhang, C. Li, and H. Wu, “A lightweight convolutional network with gated mechanism for real-time traffic anomaly detection,” IEEE Transactions on Dependable and Secure Computing, early access, 2023.

A. Patel and R. Kumar, “Neural architecture search for lightweight intrusion detection models on edge devices,” ACM Transactions on Internet of Things, vol. 4, no. 1, pp. 1–24, Feb. 2023.

H. Chen, Y. Zhang, and M. Liu, “A federated meta-learning framework for fast adaptation to dynamic attack patterns in IoT,” IEEE Transactions on Information Forensics and Security, vol. 18, pp. 120–135, 2023.

Z. Chen, L. Wang, and Q. Yang, “Granger causality meets prototype learning for unknown threat detection in IoT,” IEEE Internet of Things Journal, vol. 10, no. 5, pp. 4320–4331, Mar. 2023.

N. Gao, R. Gao, and Q. Wang, “EIW-AGTrans: An edge-implicit weighted aggregated graph Transformer for imbalanced IoT intrusion detection,” IEEE Transactions on Network and Service Management, early access, 2023.

M. H. Bhuyan, D. K. Bhattacharyya, and J. K. Kalita, “Network anomaly detection: Methods, systems and tools,” IEEE Communications Surveys & Tutorials, vol. 16, no. 1, pp. 303–336, First Quarter 2014.

Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, May 2015.

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

L. C. Molina, L. Belanche, and À. Nebot, “Feature selection algorithms: A survey and experimental evaluation,” in Proceedings of the 2002 IEEE International Conference on Data Mining (ICDM'02), 2002, pp. 306–313.

J. Li, K. Cheng, S. Wang, F. Morstatter, R. P. Trevino, J. Tang, and H. Liu, “Feature selection: A data perspective,” ACM Computing Surveys, vol. 50, no. 6, pp. 1–45, Jan. 2018.

N. Kitaev, Ł. Kaiser, and A. Levskaya, “Reformer: The efficient transformer,” arXiv preprint arXiv:2001.04451, 2020.

K. Cho, B. van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio, “Learning phrase representations using RNN encoder-decoder for statistical machine translation,” arXiv preprint arXiv:1406.1078, 2014.

D. H. Wolpert, “Stacked generalization,” Neural Networks, vol. 5, no. 2, pp. 241–259, 1992.

A. Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow. O'Reilly Media, 2022. (Chapter on Ensemble Methods)

J. Su, V. Vasudevan, A. V. D. Oord, and O. Vinyals, “Efficient transformers: A survey,” ACM Computing Surveys, vol. 55, no. 6, pp. 1–28, Dec. 2022.

DOI: http://dx.doi.org/10.26549/met.v9i1.35175