Deep-efficient-guard: securing wireless ad hoc networks via graph neural network

This study presents a new intrusion detection system (IDS) for Wireless Ad hoc Networks, leveraging graph neural networks (GNN). Overcoming the challenges faced by traditional IDS in dynamic environments, our approach focuses on meticulous feature engineering and normalization for optimal efficiency. The proposed GNN based IDS, strategically placed at zone heads, effectively filters malicious packets while considering the limited resources of wireless devices. Evaluation on the NSL-KDD dataset demonstrates the GNN-based IDS's superiority over other models like Convolutional Neural Networks (CNNs) and Transformers, highlighting its adaptability to dynamic network environments. The GNN-based IDS excels in understanding complex dependencies, contributing to efficient intrusion detection and network security in resource-constrained Wireless Ad hoc environments. Emphasizing resource efficiency, our GNN-based approach proves practical for real-time intrusion detection in resource-constrained Wireless Ad hoc networks. This work not only emphasizes GNNs' transformative role in enhancing Wireless Ad hoc network security but also contributes to discussions on efficient intrusion detection, offering innovative solutions against emerging threats. © The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management 2024. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.