Skip to content
Conference Open access

Graph Neural Network-Based Routing Optimization for Large-Scale Wireless Sensor Networks

Sep 2026 · 2026 International Conference on Secure IoT and Cybersecurity (ICOSICS) · pp. 1323-1327 · 0 citations · 10 references

Abstract

The efficiency of routing in large-scale wireless sensor networks (WSNs) is key to extend the life of a network and ensure reliable data transmission under very restrictive conditions in terms of resources. Traditional routing protocols usually suffer from the problem of scaling and adapting to the changing conditions of the network environment. Therefore, they do not perform optimally. In this paper, we present a new framework for routing optimization based on Graph Neural Networks (GNNs). This new approach inherently models the graph structure of the network to capture both spatial and temporal relationships between sensor nodes. We leverage multiple layers of graph convolutional neural networks (GCNs) to learn node embeddings that inform our adaptive and energy-aware routing decisions. Our results demonstrate through extensive simulation studies over large scale WSN topologies that our GNN-based routing approach performs much better than traditional methods by increasing the time it takes for the network to deplete its energy reserves, improving packet delivery ratios and decreasing latency, thus validating its potential for being used in scalable and robust WSN operations.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.