Abstract—Mobile Ad Hoc Networks (MANETs) are highly dynamic and decentralized, making them vulnerable to security threats such as denial-of-service (DoS) attacks, black hole attacks, and spoofing. Traditional Intrusion Detection Systems (IDS), which rely on signature-based methods, struggle to adapt to rapid topology changes and evolving attack patterns, resulting in lower detection performance. This paper proposes AI-driven IDS that employs Machine Learning (ML) and Deep Learning (DL) techniques to enhance intrusion detection in MANET environments. The proposed system utilizes anomaly detection models trained on real-time network traffic data to effectively identify and mitigate security threats. Experimental results demonstrate that the AI-driven IDS significantly outperforms traditional IDS, achieving over 95% detection accuracy compared to 85% in conventional systems, while reducing the false positive rate to less than 5%, as opposed to over 20% in traditional IDS. Additionally, the proposed system maintains a high true positive rate (above 90%), demonstrating superior detection capabilities over traditional methods, which range between 60–80%. The AI-driven IDS also offers real-time detection and mitigation, providing rapid threat response, whereas traditional IDS exhibit delayed responses in dynamic environments. Furthermore, the system adapts effectively to topology changes, ensuring continuous security in highly fluid MANET deployments. These findings confirm that the proposed AI-driven IDS significantly enhance detection accuracy, real-time adaptability, and response efficiency, making it a promising solution for securing MANETs in resource-constrained and dynamically evolving network environments.

Keywords—Mobile Ad Hoc Networks (MANETs), Intrusion Detection Systems (IDS), Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL), Anomaly Detection, Real-Time Security, Detection Accuracy, False positive rate, dynamic topology, attack mitigation, network security, federated learning, resource efficiency, mobile nodes


Cite: S. Hemalatha, K. V. S. V. Trinadh Reddy, Ramaswamy T., R. V. V. Krishna, P. Supriya, and S. N. Ananthi, “Enhancing MANET Security Using AI-Driven Intrusion Detection Systems," Journal of Communications, vol. 20, no. 4, pp. 371-383, 2025.

Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).