ML-Based Traffic Anomaly Detection and Self-Protection Framework in 5G Private Networks
Sheng-Ho Chang, Ching-Chieh Huang · 2025
As 5G private networks continue to evolve, accurately predicting traffic patterns and detecting anomalies is essential for ensuring network security and stability. This paper proposes a deep learning-driven framework that leverages Long Short-Term Memory (LSTM) networks to forecast traffic and identify abnormal behaviors in 5G subscribers. By integrating anomaly detection with self-protection framework, the system proactively blocks compromised subscribers, preventing potential threats from impacting the broader 5 G environment. Experimental results show that this approach enhances network resilience, enabling a more secure and intelligent autonomous management system for private $\mathbf{5 G}$ networks.