Cover
Vol. 2 No. 2 (2026)

Published: October 7, 2026

Pages: 94 -106

Original Article

Collaborative Lightweight Deep Learning Optimization for Intrusion Detection in SDN-IoT Networks

Ali H. Hamad 1 Mustafa Saleh 2
1 Department of Biomedical Application, College of Artificial Intelligence, University of Baghdad, Baghdad, Iraq
2 Informatics Institute for Postgraduate Studies, University of Information and Communications Technology, Baghdad, Iraq
History
  • Received: March 18, 2026
  • Revised: May 9, 2026
  • Accepted: June 1, 2026
  • Available Online: October 7, 2026
DOI

https://doi.org/10.33971/ijmrai.2.2.1

Abstract

The combination of the Internet of Things (IoT) and Software-Defined Networking (SDN) has quickly revolutionized the world of industry, and also brought with it some severe security risks. Despite its recent success, Deep learning (DL) can be a powerful method for Network Intrusion Detection systems (NIDS), but often the computational and memory requirements are too high for edge devices. In this paper, a collaborative optimization framework with lightweight deep-learning IDS models suitable for SDN-IoT environments is proposed. We formalize and test three collaborative compression schemes based on combining pruning, clustering, and quantization, yet maintaining the structural advantages of each stage: Sparsity-Preserving Quantization-Aware Training (PQAT), Cluster-Preserving Quantization-Aware Training (CQAT), and Sparsity-and-Cluster-Preserving Quantization-Aware Training (PCQAT). The three schemes are evaluated on four canonical architectures (LSTM, RNN, DNN and CNN) on the proposed SICA dataset with respect to DDoS, Probe and Brute-Force (BFA).The results show that aggressive single-stage compression such as clustering can shrink model size by up to 94%, but at the cost of predictive confidence; PCQAT, by contrast, acts as a stabilizer that recovers this lost confidence. In particular, CNN-PCQAT and DNN-PCQAT deliver the most favorable trade-off, sustaining high detection accuracy on all three attack classes with inference latency below 0.01 ms. The framework therefore offers a practical, deployment-oriented guideline for implementing real-time IDS on resource-constrained SDN-IoT edge devices.

Keywords

References

  1. Z. Wang, H. Chen, S. Yang, X. Luo, D. Li, and J. Wang, “A lightweight intrusion detection method for IoT based on deep learning and dynamic quantization”, PeerJ Computer Science, 9, 2003. https://doi.org/10.7717/peerj-cs.1569
  2. H. Dhirar and A. Hamad,” Comparative evaluation of a novel IDS dataset for SDN-IoT using deep learning models against InSDN, BoT-IoT, and ToN-IoT”, Measurement Digitalization, vol. 4, 100015, 2025. https://doi.org/10.1016/j.meadig.2025.100015
  3. “The bot-IoT dataset | UNSW Research.” Accessed: May 27, 2025. [Online]. Available: https://research.unsw.edu.au/projects/bot-iot-dataset.
  4. “ToN_IoT datasets | IEEE DataPort.” Accessed: Nov. 24, 2024. [Online]. Available: https://ieee-dataport.org/documents/toniot-datasets
  5. M. S. Elsayed, N. A. Le-Khac, A. D. Jurcut, “InSDN: a novel SDN intrusion dataset”, IEEE Access, vol. 8, 2020. https://doi.org/10.1109/ACCESS.2020.3022633
  6. H. Dhirar, and A. Hamad, “Internet of Things Software-Defined Network Intrusion Detection Dataset: Inter- and Intra-Domain”, Al-Khwarizmi Engineering Journal, vol. 21, no.3, pp. 35–47, 2025. https://doi.org/10.22153/kej.2025.08.001
  7. M. Fatima, O. Rehman and I. M. H. Rehman, "Li-IDS: An Approach Towards a Lightweight IDS for Resource-Constrained IoT," 2023 International Conference on Smart Applications, Communications and Networking (SmartNets), Istanbul, Turkey, 2023, pp. 1-6, doi: 10.1109/SmartNets58706.2023.10216096
  8. M. Lei, X. Li, B. Cai, Y. Li, L. Liu and W. Kong, "P-DNN: An Effective Intrusion Detection Method based on Pruning Deep Neural Network," 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow, UK, 2020, pp. 1-9, doi: 10.1109/IJCNN48605.2020.9206805
  9. T. Liang, J. Glossner, L. Wang, S. Shi, and X. Zhang, “Pruning and quantization for deep neural network acceleration: A survey”, In Preprint Submitted to Elsevier, vol.461, pp. 370-403, 2021. doi:10.1016/j.neucom.2021.07.045
  10. S. Sadhwani, B. Manibalan, R. Muthalagu, and P. Pawar, “A lightweight model for DDOS attack detection using machine learning techniques”. Appl. Sci., vol. 13, issue 17, 2023. https://doi.org/10.3390/app13179937
  11. B. S. Sharmila, and R. Nagapadma, “Quantized autoencoder (QAE) intrusion detection system for anomaly detection in resource-constrained IoT devices using RT-IoT2022 dataset”. Cybersecurity, vol. 6, 2023. https://doi.org/10.1186/s42400-023-00178-5
  12. A. Bouayad, H. Alami, M. J. Idrissi, I. Berrada, “College of Computing, Mohammed VI Polytechnic University (UM6P), Laboratory of Informatics, Signals, Automatics, and Cognitivism (LISAC), Faculty of Sciences Dhar El Mahraz, Sidi Mohammed Ben Abdellah University, & Abdelhak Bouayad. (2024). Lightweight federated learning for efficient network intrusion detection. In S. D. Campbell (Ed.)”, ACCESS, Vol. 12, p. 172027, https://doi.org/10.1109/ACCESS.2024.3494057
  13. R. Yu, L. L. Jeune, and N. Mentens, “University of Florida, KU Leuven, & Worcester Polytechnic Institute. (2024). Quantization-aware neural architectural search for intrusion detection”. arXiv. https://arxiv.org/pdf/2311.04194
  14. A. Rangsikunpum, S. Amiri, and L. Ost, “Ultra-Lightweight and highly efficient pruned binarised neural networks for intrusion detection in In-Vehicle networks”. Electronics, vol. 14, no. 9, p. 1710, 2025. https://doi.org/10.3390/electronics14091710
  15. S. Makenali, B. Rokh, and A. Azarpeyvand, “Integrating Pruning with Quantization for Efficient Deep Neural Networks Compression “, arXiv, 2025. https://arxiv.org/abs/2509.04244v1
  16. A. Khraisat, I. Gondal, P. Vamplew, and J. Kamruzzaman, “Survey of intrusion detection systems: techniques, datasets and challenges”, Cybersecurity, vol. 2, no. 1, pp. 1–22, 2019, doi: 10.1186/S42400-019-0038-7/FIGURES/8
  17. S. Han, H. Mao, w. J. Dally, “Deep compression: compressing deep neural networks with pruning, trained quantization and huffman coding”. In iclr, 2016. https://arxiv.org/pdf/1510.00149
  18. B. Jacob, S. Kligys, B. Chen, et al., "Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference," 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018, pp. 2704-2713, doi: 10.1109/CVPR.2018.00286
  19. M. Zhang, Y. Ma, J. Li, and J. Zhang, "A density connection weight-based clustering approach for dataset with density-sparse region", Expert Systems with Applications, vol. 230, 2023. https://doi.org/10.1016/j.eswa.2023.120633
  20. H. Hussain, P. S. Tamizharasan, and P. K. Yadav, “LCRM: Layer-wise complexity reduction method for CNN model optimization on end devices”, IEEE Access, 11, pp. 66838–66857, 2023. https://doi.org/10.1109/access.2023.3290620
  21. B. Predić, U. Vukić, M. Saračević, D. Karabašević, and D. Stanujkić, “The possibility of combining and implementing deep neural network compression methods”, Axioms, vol. 11, no. 5, 2022. https://doi.org/10.3390/axioms11050229
  22. "Collaborative Optimization." TensorFlow, Google, www.tensorflow.org/model_optimization/guide/combine/collaborative_optimization. Accessed 27 Feb. 2026.