Citation :

Ritu Bhardwaj, Ashwani Kush, "Disaster Severity Weighted GCN-LSTM Routing Framework in Underwater Sensor Networks," International Journal of Electrical and Electronics Engineering, vol. 13, no. 9, pp. 174-187, 2026. Crossref, https://doi.org/10.14445/23488379/IJEEE-V13I9P114

Abstract

The events such as earthquakes, tsunamis and underwater landslides occur; communication reliability becomes a big challenge in Underwater Wireless Sensor Networks (UWSNs). These events lead to a fast change of channels, failure of links, loss of packets and higher power usage, which diminishes the efficiency of the conventional routing protocols. Most of the protocols that are available today are based on the existing network information and do not anticipate link failure in advance. In this context, a novel Graph Convolutional Network-Long Short-Term Memory-Disaster Severity Weighting (GCN-LSTM-DSW) framework is proposed in this work to support reliable and energy-efficient routing in the presence of disasters. The GCN represents spatial relationships between the neighboring sensor nodes, and the LSTM is used to model temporal variations of the RSSI, SNR, traffic load and residual energy. The Disaster Severity Weighting (DSW) mechanism dynamically adjusts routing costs based on the prediction of network conditions and the severity of the disaster so as to identify unstable links early and select suitable paths. The proposed framework was compared with the individual GCN and LSTM models. GCN achieved an MAE of 0.071, RMSE of 0.099, and R² of 0.917, while LSTM achieved an MAE of 0.086, RMSE of 0.121, and R² of 0.891. The performance of GCN-LSTM-DSW was the best with MAE of 0.048, RMSE of 0.067, and R² of 0.958. In the synthetic disaster scenarios, the proposed framework achieved PDR values ranging from 94.0% to 97.5% across different severity levels, an average end-to-end delay of 185 ms, average energy consumption of 0.82 J, a 12.6% improvement in network lifetime, and a 34.8% reduction in route recovery time. The results show the reliability, energy efficiency and resilience enhancement in dynamic underwater disaster environments.

Keywords

Underwater sensor networks, Disaster-resilient communication, Energy-efficient routing, LSTM, Acoustic wireless networks.

References

1. Ian Fuat Akyìldìz, Dario Pompili, and Tommaso Melodia, “Underwater Acoustic Sensor Networks: Research Challenges,” Ad Hoc Networks, vol. 3, no. 3, pp. 257-279, 2005.
   [CrossRef] [Google Scholar] [Publisher Link]
2. John Heidemann et al., “Research Challenges and Applications for Underwater Sensor Networking,” IEEE Wireless Communications and Networking Conference, Las Vegas, NV, pp. 228-235, 2006.
   [CrossRef] [Google Scholar] [Publisher Link]
3. Inam Ullah et al., “Localization and Detection of Targets in Underwater Wireless Sensor using Distance and Angle based Algorithms,” IEEE Access, vol. 7, pp. 45693-45704, 2019.
   [CrossRef] [Google Scholar] [Publisher Link]
4. Nadeem Javaid et al., “An Efficient Data-Gathering Routing Protocol for Underwater Wireless Sensor Networks,” Sensors, vol. 15, no. 11, pp. 29149-29181, 2015.
   [CrossRef] [Google Scholar] [Publisher Link]
5. Radek Šalom et al., “Implementation of AODV Routing Protocol in Sensor Wireless Networks,” 2012 20^th Telecommunications Forum, Belgrade, Serbia, pp. 194-197, 2012.
   [CrossRef] [Google Scholar] [Publisher Link]
6. Mu Tong, and Minghao Tang, “LEACH-B: An Improved LEACH Protocol for Wireless Sensor Network,” 2010 6^th International Conference on Wireless Communications Networking and Mobile Computing, Chengdu, China, pp. 1-4, 2010.
   [CrossRef] [Google Scholar] [Publisher Link]
7. Safia Gul, Sana Hoor Jokhio, and Imran Ali Jokhio, “Light-Weight Depth-based Routing for Underwater Wireless Sensor Network,” 2018 International Conference on Advancements in Computational Sciences, Lahore, Pakistan, pp. 1-7, 2018.
   [CrossRef] [Google Scholar] [Publisher Link]
8. Dorsela Venkata Rami Reddy et al., “Graph Convolutional Network-based Model for Attack Detection and Mitigation Technique in Wireless Sensor Networks,” 2025 3^rd World Conference on Communication and Computing, India, pp. 1-6, 2025.
   [CrossRef] [Google Scholar] [Publisher Link]
9. S. Rajasoundaran et al., “Secure and Optimized Intrusion Detection Scheme using LSTM-MAC Principles for Underwater Wireless Sensor Networks,” Wireless Networks, vol. 30, no. 1, pp. 209-231, 2024.
   [CrossRef] [Google Scholar] [Publisher Link]
10. Mohammed F. Suleiman, and Usman Adeel, “Energy-Efficient Routing using LSTM-based Deep Learning for Sink Mobility Prediction to Enhance Lifetime and Stability of Wireless Sensor Networks,” 2023 International Symposium on Networks, Computers and Communications, Doha, Qatar, pp. 1-8, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
11. M. Shwetha, and Sannathammegowda Krishnaveni, “A Systematic Analysis, Outstanding Challenges, and Future Prospects for Routing Protocols and Machine Learning Algorithms in Underwater Wireless Acoustic Sensor Networks,” Journal of Interconnection Networks, vol. 25, no. 1, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
12. Alla Ahmad Hassan et al., “Addressing Missing Data in Machine Learning: A Comparative Insights on Imputation Techniques and Classification Accuracy,” 2025 International Conference on Next Generation Information System Engineering, Ghaziabad, Delhi (NCR), India, pp. 1-10, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
13. Rajib Kumar Halder et al., “Enhancing K-Nearest Neighbor Algorithm: A Comprehensive Review and Performance Analysis of Modifications,” Journal of Big Data, vol. 11, no. 1, pp. 1-55, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
14. Cynthia P. Haanappel, and Anne F. Voor, “Using the Interquartile Range in Infection Prevention and Control Research,” Infection Prevention in Practice, vol. 6, no. 1, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
15. Kelsy Cabello-Solorzano et al., “The Impact of Data Normalization on the Accuracy of Machine Learning Algorithms: A Comparative Analysis,” 18^th International Conference on Soft Computing Models in Industrial and Environmental Applications, Springer, Cham, pp. 344-353, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
16. Xinzheng Xu et al., “A Comprehensive Review of Graph Convolutional Networks: Approaches and Applications,” Electronic Research Archive, vol. 31, no. 7, pp. 4185-4215, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
17. Jérôme Kunegis et al., “Spectral Analysis of Signed Graphs for Clustering, Prediction and Visualization,” Proceedings of the 2010 SIAM International Conference on Data Mining, pp. 559-570, 2010.
    [CrossRef] [Google Scholar] [Publisher Link]
18. Marco Esposito et al., “Recent Advances in Internet of Things Solutions for Early Warning Systems: A Review,” Sensors, vol. 22, no. 6, pp. 1-39, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
19. Jawadul H. Bappy et al., “Hybrid LSTM and Encoder–Decoder Architecture for Detection of Image Forgeries,” IEEE Transactions on Image Processing, vol. 28, no. 7, pp. 3286-3300, 2019.
    [CrossRef] [Google Scholar] [Publisher Link]
20. Sajid Ullah Khan et al., “Energy-Efficient Routing Protocols for UWSNs: A Comprehensive Review of Taxonomy, Challenges, Opportunities, Future Research Directions, and Machine Learning Perspectives,” Journal of King Saud University Computer and Information Sciences, vol. 36, no. 7, pp. 1-23, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]