AI-Driven Digital Twins: Optimizing 5G/6G Network Slicing with NTNs

TL;DR

The paper tackles latency-sensitive eMBB resource allocation in 5G/6G NTN by coupling a digital twin (DT) with a deep deterministic policy gradient (DDPG)–based DRL agent to dynamically allocate bandwidth across UEs. The proposed DT-DRL framework uses real-time DT synchronization and continuous-action DRL to model and adapt to NTN dynamics, including FBS mobility and blockages, to minimize latency and optimize resource usage. Key contributions include real-time DT state synchronization, a continuous-action DRL policy for bandwidth allocation, and NTN-aware optimization that yields substantial latency reductions (approximately 25 percent over static methods) and higher resource utilization in disaster or urban-obstruction scenarios. The work demonstrates a scalable approach for 5G/6G NTN network slicing with practical impact on disaster response and dense urban environments, while outlining avenues for extending to multi-slice and energy-efficient URLLC/mMTC deployments.

Abstract

Network slicing in 5G/6G Non-Terrestrial Network (NTN) is confronted with mobility and traffic variability. An artificial intelligence (AI)-based digital twin (DT) architecture with deep reinforcement learning (DRL) using Deep deterministic policy gradient (DDPG) is proposed for dynamic optimization of resource allocation. DT virtualizes network states to enable predictive analysis, while DRL changes bandwidth for eMBB slice. Simulations show a 25\% latency reduction compared to static methods, with enhanced resource utilization. This scalable solution supports 5G/6G NTN applications like disaster recovery and urban blockage.

Figures (3)

• Figure 1: System Model: Network architecture with gBS, FBS, UEs and DT interactions.
• Figure 2: Proposed AI-driven DI architecture.
• Figure 3: Performance comparison of proposed method with baseline techniques for eMBB slice.