Revolutionizing Network Autonomy with Intent-Based Network Digital Twins: Architecture, Challenges, and Innovations

As sixth-generation (6G) wireless networks evolve, it is clear that the increasing diversity of network services and architectures is making network management significantly more complex. Autonomous networks have emerged to address this challenge, aiming to deliver intelligent, large-scale autonomy and on-demand services across various environments. However, new applications have introduced more varied and stringent requirements in terms of latency, transmission speed, and reliability. User preferences and device diversity further complicate the situation. Seamless cross-service compatibility in intelligent, automated network management is now more challenging than ever. Existing management methods simply cannot meet the demands of intelligent autonomy in such diverse scenarios. We propose an intent-based network digital twin (IBN-DT) management framework to address these issues. This architecture separates the knowledge and control planes, incorporating a multi-level, knowledge-driven intent processing system. By utilizing network digital twins (NDT), the framework enables efficient simulation, policy deployment, and monitoring, creating a fully autonomous management loop. This article outlines the complete process from intent to policy execution and provides detailed use cases and experimental results. In resource-constrained scenarios, our algorithm has demonstrated a 40% improvement in revenue.