Demonstrates a modular user-plane architecture that enhances automation in next-generation mobile networks, indicating significant improvements in flexibility.
This paper presents a modular user-plane architecture for next-generation mobile core networks that addresses the limitations of the monolithic User Plane Function (UPF) in 5G systems. The proposed design decomposes the UPF into independently deployable modules, enabling flexible composition and customizable user-plane paths. To support this architecture, the paper introduces a control-plane framework that enables the selection, chaining, and configuration of modular UPF components while remaining aligned with existing 5G core procedures. In addition, a multi-agent AI-based system is proposed to automate the creation of user-plane paths from high-level operator intents, with specialized agents handling intent interpretation, module selection, configuration, and deployment. The approach is evaluated across multiple intent-driven scenarios with varying complexity using different large language models. The results demonstrate effective end-to-end operation and highlight the benefits of combining modular user-plane design with agent-based automation for flexible and programmable next-generation networks.
No takes yet. Share an insight, caveat, or question.
Aintablian et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: