This paper is published in Volume-11, Issue-1, 2025
Area
Computer Engineering
Author
Ibraheem Kateeb, Yasser Ahmed
Org/Univ
Qassim University, Saudi Arabia, Saudi Arabia
Pub. Date
29 January, 2025
Paper ID
V11I1-1217
Publisher
Keywords
Software Defined Networks (SDN), Fog computing, Anomaly Detection, Generative Adversarial Networks (GANs), Gated Recurrent Unit (GRU), Wireless Networks, Smart Cities.

Citationsacebook

IEEE
Ibraheem Kateeb, Yasser Ahmed. Anomaly Detection in SDN-enabled Cloud-Fog Collaborative Networks, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
Ibraheem Kateeb, Yasser Ahmed (2025). Anomaly Detection in SDN-enabled Cloud-Fog Collaborative Networks. International Journal of Advance Research, Ideas and Innovations in Technology, 11(1) www.IJARIIT.com.

MLA
Ibraheem Kateeb, Yasser Ahmed. "Anomaly Detection in SDN-enabled Cloud-Fog Collaborative Networks." International Journal of Advance Research, Ideas and Innovations in Technology 11.1 (2025). www.IJARIIT.com.

Abstract

Software-defined networking (SDN) has brought about a paradigm shift in network management and control, offering increased flexibility and automation capabilities. This transformation is particularly relevant in the context of smart cities, where integrating IoT devices and smartphones is essential for delivering efficient and responsive city services. As SDN gains prominence, the security of these devices in fog computing environments becomes a critical concern for maintaining the integrity and reliability of smart city infrastructures. Effective access control mechanisms are needed to safeguard the network, tailored to the unique requirements of SDN networks. These mechanisms are crucial for smart cities, where many devices and systems need to interact seamlessly while maintaining high-security standards. Additionally, KPI anomaly detection in SDN networks poses challenges due to the real-time processing of copious data. To address these challenges, this paper proposes a cloud-fog collaborative architecture with a GAN-GRU model for malicious activity detection. This architecture is designed with smart cities in mind, where fog nodes divide the network into subregions, mirroring the geographical divisions of a city. The cloud is responsible for model training, while fog nodes execute detection tasks, ensuring a responsive and efficient system for anomaly detection. The proposed method outperforms benchmark algorithms in terms of precision, recall, and F1 score, demonstrating its potential for implementation in smart city infrastructures. Furthermore, the impact of time window length on anomaly detection is analyzed, revealing optimal performance with a window length of 70. This paper also introduces a reputation-based access restriction management mechanism, demonstrating its effectiveness in preventing unauthorized access while ensuring secure operations.