Smart City Gnosys
Smart city article details
| Title | Enhancing Cyber Security In 6G Networks With Federated Learning For Collaborative Threat Mitigation |
|---|---|
| ID_Doc | 23762 |
| Authors | Gupta S.; Roy D.; Pramanik P.; Chowdhury S. |
| Year | 2025 |
| Published | Advanced Sciences and Technologies for Security Applications, Part F458 |
| DOI | http://dx.doi.org/10.1007/978-3-031-85008-0_11 |
| Abstract | 6G networks are a significant step forward in the development of wireless communication systems, which offer improved interconnectivity, higher speeds, and broader applications, including IoT, autonomous systems, and smart cities. At the same time, the features of 6G networks, such as their distributed structure, enormous machine-to-machine interactions, and high volume of fluctuating information flow, pose unique challenges to cybersecurity. Most conventional perimeter security measures would be ill-suited to address the problems presented by these vast networks, underscoring the need for novel, cost-effective, and robust responses. Federated Learning (FL) has gained attention as a promising technique in improving cybersecurity of 6G networks. In contrast with the traditional frameworks, FL allows for data to remain on the device enabling devices to work together to carry out the data pre-processing within the device without the need to transmit so much information. Nevertheless, implementing FL in 6G networks has its issues. Among them are lack of uniform information distribution, issues around the inclusion of swapping devices that are low powered and readily accessible, and wide ranging across billions of devices. These constraints imply that a framework addressing 6G ecosystem needs to be designed. This study proposes a new Federated Learning-based cyber security model that seeks to conform to the specific characteristics of 6G networks without being limited by classical systems. Other innovations consist in adaptive aggregation such that updates from devices located in hostile environments are prioritized for model updates. This ensures that the protective model is fine-tuned to potential security threats. Active engagement underscores device utilization anchored upon computational, network, and threat potential limits (resource saving across diverse environments is employed). Moreover, devices contribute to mitigating power range through unmasking TIS—threat intelligence sharing—which aids perception of higher order threats with devices sharing extracts related to abnormalities observed at a higher level than their own, thereby bringing more to the table than just contributions to local models. The architecture also recognizes egoistic aspirations of 5G network slices extending from IoT network to emergency network. These tailor made models improve precision in model working by serving the necessary security and operational criteria for all the slices and offer high availability for various services particularly in eHealth and automatic systems. In addition, various techniques such as differential privacy are implemented in order to protect sensitive information without compromising the usefulness of the model of a particular scenario and satisfy security and efficiency measures simultaneously. A clear performance upper hand has been established through several of evaluations of the proposed paradigm. Threat detection has markedly advanced due to the adaptive aggregation and the TIS layer, already overcoming conventional FL and centralized learning reaches. The structural formation of the framework resolved the issue of scalability and hence low communication delay was achievable in large networks. Personalized slice-specific slices outperformed generalized ones, especially in healthcare where accuracy of diagnosis is core measurement of the applied systems. This study highlights the salient role FL has in improving cyber security within next generation networks. By addressing important concerns such as scalability, privacy and adaptability, the framework sets a solid basis for the protection of the 6G system against sophisticated threats. Further research will focus on managing energy optimization for resource efficient scenarios. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. |
| Author Keywords | 6G Networks; Cybersecurity framework; Federated Learning (FL); Privacy-preserving mechanisms; Threat Intelligence Sharing (TIS) |
Similar Articles
| Id | Similarity | Authors | Title | Published |
|---|---|---|---|---|
26334  | 0.922 | Sirohi D.; Kumar N.; Rana P.S.; Tanwar S.; Iqbal R.; Hijjii M. | Federated Learning For 6G-Enabled Secure Communication Systems: A Comprehensive Survey | Artificial Intelligence Review, 56, 10 (2023) |
| 57646 | 0.911 | Kholidy H.A.; Hariri S. | Toward An Experimental Federated 6G Testbed: A Federated Leaning Approach | Proceedings of IEEE/ACS International Conference on Computer Systems and Applications, AICCSA, 2022-December (2022) |
| 19206 | 0.88 | Thakur A.; Tyagi R.; Tripathy H.K.; Yang T.; Rathore R.S.; Mo D.; Wang L. | Detecting Network Attack Using Federated Learning For Iot Devices | International Conference on Intelligent Algorithms for Computational Intelligence Systems, IACIS 2024 (2024) |
| 6200 | 0.873 | Shen S.; Yu C.; Zhang K.; Ni J.; Ci S. | Adaptive And Dynamic Security In Ai-Empowered 6G: From An Energy Efficiency Perspective | IEEE Communications Standards Magazine, 5, 3 (2021) |
| 23730 | 0.872 | Sharma A.; Rani S. | Enhancing 6G-Iot Network Security: A Trustworthy And Responsible Ai-Driven Stacked-Hybrid Model For Attack Detection | IEEE Internet of Things Journal (2025) |
| 47763 | 0.871 | Alotaibi A.; Barnawi A. | Securing Massive Iot In 6G: Recent Solutions, Architectures, Future Directions | Internet of Things (Netherlands), 22 (2023) |
| 58290 | 0.87 | Yi W.; Fu Y.; Cao J.; Gan L.; Xiong L.; Li H. | Towards Seamless 6G And Ai/Ml Convergence: Architectural Enhancements And Security Challenges | IEEE Network (2025) |
| 22310 | 0.869 | Adjewa F.; Esseghir M.; Merghem-Boulahia L. | Efficient Federated Intrusion Detection In 5G Ecosystem Using Optimized Bert-Based Model | International Conference on Wireless and Mobile Computing, Networking and Communications (2024) |
| 26342 | 0.862 | Quy V.K.; Nguyen D.C.; Van Anh D.; Quy N.M. | Federated Learning For Green And Sustainable 6G Iiot Applications | Internet of Things (Netherlands), 25 (2024) |
| 54221 | 0.859 | Yang H.; Liu H.; Yuan X.; Wu K.; Ni W.; Zhang J.A.; Liu R.P. | Synergizing Intelligence And Privacy: A Review Of Integrating Internet Of Things, Large Language Models, And Federated Learning In Advanced Networked Systems | Applied Sciences (Switzerland), 15, 12 (2025) |