Copyright © 2026 Authors retain the copyright of this article. This article 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.

@article{174475,
        author = {B. S. Kiruthika Devi and Sivesh Shrutik Muhilvannan and Nishanth Babu H},
        title = {Federated Learning Empowered by Attention Augmented CNN for Edge-Based Intrusion  Detection},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {10},
        pages = {4429-4436},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=174475},
        abstract = {In today’s swiftly developing technological environment, where iot and edge computing is becoming vaster and more dominant. The rapid growth introduces a variety of concerns regarding different cyber threats and attacks which are focused on edge computing networks and its devices. The increasing number of interconnected devices in the edge network would proportionally have a large amount of data flowing through it on a daily basis. This exposes itself to the constraints of the conventional Intrusion Detection System (IDS). The data collection process under Traditional and Centralized Intrusion Detection Systems occurs through a solitary server that handles data acquisition from edge devices which leads to major privacy problems and security weaknesses. The centralized models are not best suited for edge environments where computational resources are constrained. Federated learning permits the models to be trained across the decentralized devices without the need to share their raw data to themselves or to the server. The proposed framework aggregates the updates from each edge device and the models trained by them to create a global IDS model to ensure security and efficiency at a global range. The data used for this framework is simulated using the Edge-IIoTset Cyber Security Dataset of IoT & IIoT dataset. This approach is evaluated by detection accuracy, model convergence and impact of federated aggregation on performance. The results validate that validate federated learning as a privacy-preserving IDS solution. The successful implementation of the system could considerably enhance the security of edge computing environment, making it more efficient and safer.},
        keywords = {Federated Learning, Intrusion Detection Systems, Edge Computing, Privacy-Preserving, IoT Security, Edge-IIoTset Dataset, Detection Accuracy.},
        month = {March},
        }