A CNN–LSTM Hybrid Deep Learning Technique for Predicting Performance Degradation in SDH Networks

Authors

F. U. Didigwu

Department of Electrical/Electronic Engineering, University of Port Harcourt (Nigeria)

Article Information

DOI: 10.51244/IJRSI.2026.1309000101

Subject Category: Computer Science

Volume/Issue: 13/9 | Page No: 1409-1426

Publication Timeline

Submitted: 2026-09-17

Accepted: 2026-09-22

Published: 2026-10-09

Abstract

The increasing complexity of Synchronous Digital Hierarchy (SDH) networks necessitates reliable forecasting of network performance indicators to support proactive performance monitoring. This study evaluates deep learning approaches for forecasting key performance indicators (KPIs) in SDH networks using real-world Network Management System (NMS) data from Uyo, Nigeria. The dataset comprised 52,128 observations collected at 15-minute intervals from January to June 2026 across three SDH links: UYO_EKET, UYO_ITU, and UYO_ORON. Five KPIs—Bit Error Rate (BER), Errored Seconds (ES), Severely Errored Seconds (SES), Background Block Errors (BBE), and Unavailable Seconds (UAS)—were examined. Data were divided chronologically into training (January–April), validation (May), and independent testing (June) periods. Four deep learning architectures—Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and CNN–LSTM—were evaluated for one-step-ahead (15-minute) forecasting using a 24-step (6-hour) historical sequence. Standardization was fitted exclusively on the training data to prevent information leakage. Performance was assessed using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE). Results demonstrated that the CNN–LSTM model consistently produced the lowest MSE, RMSE, and MAE, while its prediction curves generally remained closer to the observed temporal behaviour than those of the individual RNN, CNN, and LSTM models. Across the KPIs, the models successfully captured the underlying temporal patterns but tended to smooth abrupt fluctuations and extreme observations, particularly in BBE, where the actual series exhibited substantial spikes. SES and UAS were predominantly concentrated around zero, resulting in relatively small absolute prediction errors despite some visible deviations in the plots. The consistency between the numerical metrics and graphical results indicates that the CNN–LSTM model provides more stable and accurate forecasting of SDH network KPIs. The findings demonstrate the potential of Hybrid Deep Learning architectures to enhance proactive network performance assessment through short-term forecasting of emerging degradation patterns, supporting the early identification of abnormal network conditions.

Keywords

Synchronous Digital Hierachy (SDH), Network Performance, Deep Learning Models, Network KPI, Metrics

Downloads

References

1. Binjumah, W. (2024) The Role of Machine Learning and Deep Learning Approaches to Improve Optical Communication Systems. Journal of Intelligent Learning Systems and Applications, 16, 418-429. doi: 10.4236/jilsa.2024.164021. [Google Scholar] [Crossref]

2. Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2020). A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 32(1), 4–24. Publisher: IEEE [Google Scholar] [Crossref]

3. Mao, Q., Hu, F., & Hao, Q. (2018). Deep learning for intelligent wireless networks: A comprehensive survey. IEEE Communications Surveys & Tutorials, 20(4), 2595–2621. [Google Scholar] [Crossref]

4. Publisher: IEEE https://doi.org/10.1109/COMST.2018.2846401 [Google Scholar] [Crossref]

5. Yu, R., Xie, S., Zhang, Y., & Gjessing, S. (2018). A deep learning approach for network traffic prediction. IEEE Transactions on Network and Service Management, 15(4), 1231–1245. [Google Scholar] [Crossref]

6. Publisher: IEEE https://doi.org/10.1109/TNSM.2018.2864244 [Google Scholar] [Crossref]

7. Lai, G., Chang, W.-C., Yang, Y., & Liu, H. (2018). Modeling long- and short-term temporal patterns with deep neural networks. Proceedings of the 41st International ACM SIGIR Conference, 95–104. Publisher: ACM https://doi.org/10.1145/3209978.3210006 [Google Scholar] [Crossref]

8. Kiranyaz, S., Ince, T., & Gabbouj, M. (2019). 1D convolutional neural networks and applications: A survey. Mechanical Systems and Signal Processing, 151, 107398. Publisher: Elsevier https://doi.org/10.1016/j.ymssp.2019.107398 [Google Scholar] [Crossref]

9. Sun, Y., Peng, M., Zhou, Y., Huang, Y., & Mao, S. (2019). Application of machine learning in wireless networks: Key techniques and open issues. IEEE Communications Surveys & Tutorials, 21(4), 3072–3108. Publisher: IEEE https://doi.org/10.1109/COMST.2019.2924243 [Google Scholar] [Crossref]

10. He, Y., Yu, F. R., Zhao, N., Yin, H., & Leung, V. C. M. (2019). Deep reinforcement learning for resource allocation in wireless networks. IEEE Wireless Communications, 26(3), 44–51. [Google Scholar] [Crossref]

11. Publisher: IEEE https://doi.org/10.1109/MWC.2019.1800298 [Google Scholar] [Crossref]

12. Wang, X., Han, Y., Wang, C., Zhao, Q., Chen, X., & Chen, M. (2020). In-edge AI: Intelligentizing mobile edge computing, caching and communication. IEEE Network, 33(5), 156–165. Publisher: IEEE https://doi.org/10.1109/MNET.001.1800286 [Google Scholar] [Crossref]

13. Lim, B., Zohren, S., & Roberts, S. (2021). Time-series forecasting with deep learning: A survey. Philosophical Transactions of the Royal Society A, 379(2194). Publisher: The Royal Society https://doi.org/10.1098/rsta.2020.0209 [Google Scholar] [Crossref]

14. Zhang, Q., Yang, L. T., Chen, Z., & Li, P. (2019). A survey on deep learning for big data. Information Fusion, 42, 146–157. Publisher: Elsevier https://doi.org/10.1016/j.inffus.2017.10.006 [Google Scholar] [Crossref]

15. Al-Fuqaha, A., Guizani, M., Mohammadi, M., Aledhari, M., & Ayyash, M. (2018). [Google Scholar] [Crossref]

16. Internet of Things: A survey on enabling technologies, protocols, and applications. [Google Scholar] [Crossref]

17. IEEE Communications Surveys & Tutorials, 17(4), 2347–2376. Publisher: IEEE https://doi.org/10.1109/COMST.2015.2444095 [Google Scholar] [Crossref]

18. Ahmed, M., Mahmood, A. N., & Hu, J. (2019). A survey of network anomaly detection techniques. Journal of Network and Computer Applications, 60, 19–31. [Google Scholar] [Crossref]

19. Publisher: Elsevier https://doi.org/10.1016/j.jnca.2015.11.016 [Google Scholar] [Crossref]

20. Sainath, T. N., Vinyals, O., Senior, A., & Sak, H. (2015). Convolutional, long short-term memory, fully connected deep neural networks. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 4580–4584. Publisher: IEEE https://doi.org/10.1109/ICASSP.2015.7178838 [Google Scholar] [Crossref]

21. Cho, K., Van Merriënboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014). Learning phrase representations using RNN encoder–decoder for statistical machine translation. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, 1724–1734. Publisher: ACL https://doi.org/10.3115/v1/D14-1179 [Google Scholar] [Crossref]

22. Zhao, Z., Chen, W., Wu, X., Chen, P. C. Y., & Liu, J. (2018). [Google Scholar] [Crossref]

23. LSTM network: A deep learning approach for short-term traffic forecast. IET Intelligent Transport Systems, 11(2), 68–75. Publisher: IET https://doi.org/10.1049/iet-its.2016.0208 [Google Scholar] [Crossref]

24. Yunfeng Ge , Zonghuan Guo , Peng Wang, Xiaogang Li, Hongyan Li (2025): Adaptive clockwork LSTM for network traffic prediction, Journal of Information and Intelligence, https://doi.org/10.1016/j.jiixd.2025.06.004 [Google Scholar] [Crossref]

25. 16. Shengxiang Lv, Lin Wang, Sirui Wang (2023): A Hybrid Neural Network Model for Short-Term Wind Speed Forecasting. https://doi.org/10.3390/en16041841 [Google Scholar] [Crossref]

26. Joanna Waczyńska, Edoardo Martelli, Sofia Vallecorsa, Edward Karavakis and Tony Cass (2021). EPJ Web Conf. Volume 251, 2021. 25th International Conference on Computing in High Energy and Nuclear Physics (CHEP 2021). DOI: https://doi.org/10.1051/epjconf/202125102050. [Google Scholar] [Crossref]

27. Rusul L. Abduljabbar, Hussein Dia, Pei-Wei Tsai (2021). Unidirectional and Bidirectional LSTM Models for Short-Term Traffic Prediction. https://doi.org/10.1155/2021/5589075 [Google Scholar] [Crossref]

28. Indrajeet Kumar, Bineet Kumar Tripathi, Anugrah Singh (2023): Attention-based LSTM network-assisted time series forecasting models for petroleum production. Engineering Applications of Artificial Intelligence. Volume 123, Part C, August 2023, 106440. https://doi.org/10.1016/j.engappai.2023.106440 [Google Scholar] [Crossref]

29. Jihong Zhao, Xiaoyuan He (2022): NTAM-LSTM: An attention-based LSTM network for network traffic prediction. MATEC Web of Conferences, 355, Article 02007. Publisher: EDP Sciences. https://doi.org/10.1051/matecconf/202235502007. [Google Scholar] [Crossref]

30. Liu, X., Li, Y., Zhu, S., Su, Q., Dai, J., Li, C., Zhu, J., & Zhang, J. (2025). [Google Scholar] [Crossref]

31. A traffic forecasting framework for cellular networks based on a dynamic component management mechanism. Electronics, 14(20), 4003. Publisher: MDPI, https://doi.org/10.3390/electronics14204003 [Google Scholar] [Crossref]

Metrics

Views & Downloads

Similar Articles