Modern 5G networks must support ultra-low latency, reliability, and rapidly changing traffic patterns, making real-time monitoring and intelligent management increasingly difficult. This article explores how Digital Twin technology can address these challenges by creating a live virtual replica of a 5G network capable of monitoring, simulation, and behavioral analysis in real time.
The presented system combines Open5GS, UERANSIM, Prometheus, and LSTM-based neural networks to classify network behavior using synthetically generated traffic data. The research focuses on one of the most critical problems in AI-driven network management: domain transfer — the gap between synthetic simulation data and real-world network conditions.
Experimental results show that while models achieve very high accuracy in simulated environments, performance drops significantly when applied to real network traffic, highlighting the limitations of current simulation fidelity and the importance of domain adaptation techniques for future 6G infrastructures.
The article also demonstrates how lightweight Digital Twins can evolve from passive monitoring systems into adaptive, intelligent platforms capable of continuous learning, fine-tuning, and autonomous optimization. These findings provide important insights into the future of self-managing 6G communication networks, semantic networking, and AI-native telecom infrastructure.

DavidDev

Digital Twin technology applied to 5G networks can create live virtual replicas for real-time monitoring, simulation, and behavioral analysis. A system combining Open5GS, UERANSIM, Prometheus, and LSTM-based neural networks classifies network behavior using synthetic traffic data. A key finding is the domain transfer problem: models trained on simulated data lose significant accuracy when applied to real network traffic, underscoring the need for domain adaptation techniques. The research outlines how lightweight Digital Twins can evolve into adaptive, self-managing platforms with continuous learning capabilities, offering a foundation for future 6G AI-native telecom infrastructure.

Digital Twins for 5G and Beyond: Real-Time Network Intelligence for the 6G Era