HIERARCHICAL EDGE-CLOUD FRAMEWORK FOR REAL-TIME NETWORK ANOMALY DETECTION AND SECURITY
DOI:
https://doi.org/10.5281/zenodo.21818022Keywords:
Hierarchical edge-cloud architecture Real-time anomaly detection Network security Distributed intrusion detection Edge intelligence Cloud computing Machine learning Cybersecurity analytics Adaptive threat monitoring IoT security.Abstract
The hundreds of proliferation in distributed digital infrastructures and certain latency-sensitive applications require serious real-time cybersecurity solutions for heterogeneous environments. Traditional centralized approach to monitoring systems have their own set of challenges including volume capability, high latency, and delayed response which is more significant for large-scale IoT and edge-enabled networks. This study proposes Hierarchical Edge-Cloud Framework for Real-Time Network Anomaly Detection and Security with multi-layered monitoring, collaborative intelligence and adaptive machine learning. Lightweight detection models that are deployed at edge nodes perform immediate traffic inspection while computationally intensive deep learning models in the cloud perform global pattern analysis and model optimization. A hierarchical coordination mechanism is used to achieve continuous knowledge sharing and adaptable updates at edge and cloud level. Experimental evaluation using large volume of network traffic data results show 98.6% accuracy in detection with 42% saving in response latency as compared to centralized systems and a 35% improvement in accuracy in classifying threats. Localized filtering reduces network bandwidth usage even further (28%). These results are useful in emphasizing the framework's capacity to provide scalable, resilient and efficient cybersecurity for next-generation distributed networks.
References
[1] Admass, Wasyihun Sema, Yirga Yayeh Munaye, and Abebe Abeshu Diro. "Cyber security: State of the art, challenges and future directions." Cyber Security and Applications 2 (2024): 100031.
[2] AlDaajeh, Saleh, and Saed Alrabaee. "Strategic cybersecurity." Computers & Security 141 (2024): 103845.
[3] Choudhury, Nobhonil Roy, and Shyamalendu Paul. "Comparative analysis of traditional vs. AI-driven network security." In Human Impact on Security and Privacy: Network and Human Security, Social Media, and Devices, pp. 53-74. IGI Global Scientific Publishing, 2025.
[4] Akinsanya, Michael Oladipo, Oluwafemi Clement Adeusi, and Kazeem Bamidele Ajanaku. "A detailed review of contemporary cyber/network security approaches and emerging challenges." Communication in Physical Sciences 8, no. 4 (2022): 721-732.
[5] Zhou, Yujie, Ruyan Wang, Xingyue Mo, Zhidu Li, and Tong Tang. "Robust hierarchical federated learning with anomaly detection in cloud-edge-end cooperation networks." Electronics 12, no. 1 (2022): 112.
[6] Zhang, Shenglin, Jiacheng Zhang, Guohua Liu, Shiqi Chen, Chenyu Zhao, Minghua Ma, Yutong Chen, Yongqian Sun, and Dan Pei. "Bridging Edge and Cloud: A Knowledge-Enhanced Framework for Efficient Time Series Anomaly Detection." IEEE Transactions on Services Computing (2025).
[7] Yang, Tao, Weijie Hao, Qiang Yang, and Wenhai Wang. "Cloud-edge coordinated traffic anomaly detection for industrial cyber-physical systems." Expert Systems with Applications 230 (2023): 120668.
[8] Jiang, Bingcheng, Qian He, Zhongyi Zhai, and Hang Su. "Anomaly Detection and Access Control for Cloud-Edge Collaboration Networks." Intelligent Automation & Soft Computing 37, no. 2 (2023).
[9] Reis, Manuel JCS, and Carlos Serôdio. "Edge AI for real-time anomaly detection in smart homes." Future Internet 17, no. 4 (2025): 179.
[10] Nawaal, Bakhtawar, Usman Haider, Inam Ullah Khan, and Muhammad Fayaz. "Signature-based intrusion detection system for IoT." In Cyber security for next-generation computing technologies, pp. 141-158. CRC Press, 2024.
[11] Khan, Muhammad Ali, Rao Naveed Bin Rais, Osman Khalid, and Fiaz Gul Khan. "A comparative analysis of federated and centralized machine learning for intrusion detection in IoT." In 2023 24th International Arab Conference on Information Technology (ACIT), pp. 1-7. IEEE, 2023.
[12] Deng, Zihao, and Geert Deconinck. "Machine Learning-Based Distributed Intrusion Detection System in Industrial Edge Environments: Challenge Identification." In 2025 20th European Dependable Computing Conference Companion Proceedings (EDCC-C), pp. 44-47. IEEE, 2025.
[13] Vani, K., and S. P. Swornambiga. "Adaptive intrusion detection framework for enhanced cloud security in fog and edge computing environments." International Journal of Advanced Technology and Engineering Exploration 11, no. 121 (2024): 1613.
[14] Thein, Thin Tharaphe, Yoshiaki Shiraishi, and Masakatu Morii. "Personalized federated learning-based intrusion detection system: Poisoning attack and defense." Future Generation Computer Systems 153 (2024): 182-192.
