A Machine Learning–Based Framework for Traffic Classification and Threshold-Based Load Management in 5G Network Slicing

Authors

• Safi Ibrahim, Department of Information Technology, Egyptian E-Learning University (EELU), Egypt
• Younis S. Younis, Department of Information Technology, Faculty of Computer and Information, South Valley University, Qena 83523, EgyptFollow
• Kamal S. Hamza, Department of Information Technology, Egyptian E-Learning University (EELU), Egypt
• Mohamed M. Ashour, EELU - College of Information Technology (CIT) and Faculty of Engineering, Mansoura University, Egypt

Subject Area

Computer Science

Article Type

Original Study

Abstract

This study proposes a machine learning–based framework that applies machine learning techniques to improve the efficiency of 5G network slicing through automated traffic classification and threshold-based load management . The proposed model optimizes resource allocation among the three standardized 5G slice types: enhanced Mobile Broadband (eMBB), ultra-Reliable Low-Latency Communication (URLLC), and massive Machine-Type Communication (mMTC). Two supervised learning algorithms—K-Nearest Neighbors (KNN) and Support Vector Machine (SVM)—are trained using Quality of Service (QoS) parameters such as packet delay, loss rate, and Quality Class Identifier (QCI). Experimental evaluations were conducted on two large-scale datasets containing over 400,000 traffic instances, demonstrating that the KNN classifier achieved a peak accuracy of 97.6%, outperforming SVM, which reached 92.8%. Additionally, a threshold-based traffic management mechanism was implemented to monitor slice utilization in real time and reroute traffic when predefined thresholds were exceeded. The results confirm that this intelligent approach effectively minimizes slice congestion, enhances throughput, and ensures continuous service delivery. The proposed system represents a scalable, data-driven solution for managing heterogeneous 5G environments, paving the way toward fully autonomous 6G-ready network architectures.

Keywords

5G, Network Slicing, Enhanced Mobile Broadband (eMBB), Massive Machine-Type Communications (mMTC), Ultra-Reliable Low-Latency Communications (URLLC), Machine-Learning (ML).

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 License.

Recommended Citation

Ibrahim, Safi; Younis, Younis S.; Hamza, Kamal S.; and Ashour, Mohamed M. (2026) "A Machine Learning–Based Framework for Traffic Classification and Threshold-Based Load Management in 5G Network Slicing," Mansoura Engineering Journal: Vol. 51 : Iss. 3 , Article 9.
Available at: https://doi.org/10.58491/2735-4202.3462