Optimizing energy efficiency in IoT enabled wireless sensor networks via Gegenbauer graph neural network
Wireless Sensor Networks (WSNs) have been widely studied because of their data collection and observation across diverse applications. However, the restricted energy resources of sensor nodes (SNs) remain a major challenge for prolonging the network lifetime. To address this issue, this study proposes an Energy Efficient IoT-Enabled WSN framework based on a Gegenbauer Graph Neural Network (EE-IoT-WSN-GGNN). Initially, clusters are generated using Multi-Kernel Collaborative Fuzzy Clustering (MKCFC) method. For each cluster, an efficient Cluster Head (CH) is selected with the help of Elite Opposite Sparrow Search algorithm (EOSSA), facilitating WSN development to accommodate a large amount of SNs. The multi objective fitness dependent optimizer algorithm (MOFDOA) then optimizes the candidate routing configurations generated by the trained GGNN to find the most energy efficient data transmission using the Adam optimizer. The proposed framework can simultaneously optimize energy consumption, delay, congestion, QoS, and network stability by combining the topology-aware routing from GGNN with multi-objective routing optimization. The proposed EE-IoT-WSN-GGNN framework was implemented in the python environment. The performance of the proposed EE-IoT-WSN-GGNN approach attains 20.65%, 25.81% and 31.21% higher energy efficiency, 21.58%, 25.15% and 30.15% higher network stability compared with existing methods like efficient energy management in WSN depending on optimized explicit feature interaction-aware graph neural network (MEE-WSN-DT), cluster-basis routing protocol design utilizing Gated Fusion Adaptive Graph Neural Network in WSN (EE-MPDP-WSN), energy efficient clustering for dense WSN using graph neural networks with coverage metrics (LMNN-EE-WSN), multi-hop graph-based approach for an energy-efficient routing protocol in wireless sensor networks (QoS-ASR-GNN), Lightweight reinforcement learning for energy efficient communications in wireless sensor networks (ONS-PO-GNN), energy-efficient optimization-based routing technique for wireless sensor network using machine learning (GAT-EER-WSN) and Graph convolutional networks and deep reinforcement learning in IoT environs for intelligent edge routing (GCN-SRL-ER) respectively
Authors
- Anita Jones Mary
- Prakash Arumugam (ORCID: https://orcid.org/0000-0002-7675-2087)
- Anand Rajendran
- Abinayaa S. S.
Institutions
- Karunya University (IN)
- Karnavati University (IN)
- Amrita Vishwa Vidyapeetham (IN)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-11-20
- DOI
- https://doi.org/10.1038/s41598-026-68368-y
- Primary Topic
- Energy Efficient Wireless Sensor Networks
- Type
- article
- Field-Weighted Citation Impact
- 0.00