AI-Powered Digital Twin for Heterogeneous Wireless Network Simulation

Abstract

To address the challenges of managing heterogeneous wireless environments (HWNs), which integrate diverse access networks like 5G, LTE, and Wi-Fi, this research introduces an AI-powered digital twin framework. The framework creates real-time virtual replicas of HWNs to enable dynamic monitoring, simulation, and optimization. Leveraging advanced AI techniques, including reinforcement learning, predictive analytics, and anomaly detection, the framework achieves significant improvements in network performance. Simulations demonstrate a 25% reduction in latency, a 20% improvement in throughput, and a 20% decrease in energy consumption compared to state-of-the-art approaches. Additionally, user-centric Quality of Service (QoS) and Quality of Experience (QoE) models ensure enhanced user satisfaction by tailoring network performance to evolving demands. The authors validated the framework's scalability and adaptability across diverse scenarios, such as dynamic handover management, energy-efficient resource allocation, and fault recovery. These results highlight the potential of the proposed framework to revolutionize network management in next-generation networks, such as 6G. Concluding, this research underscores the framework’s ability to provide real-time optimization, significant energy savings, and improved user satisfaction while addressing challenges like computational overhead. Future work includes integrating renewable energy sources, strengthening cybersecurity features, and expanding capabilities to support emerging technologies like semantic communication.