KP-A: a unified network knowledge plane for catalyzing agentic network intelligence

Abstract

The emergence of large language models (LLMs) and agentic systems is enabling autonomous 6G networks with advanced intelligence, including self-configuration, self-optimization, and self-healing. However, the current implementation of individual intelligence tasks necessitates isolated knowledge retrieval pipelines, resulting in redundant data flows and inconsistent interpretations. There is now a growing need to support greater multi-tenanted interoperability in 6G, especially for Open RAN architectures, and we propose KP-A: a unified Network Knowledge Plane specifically designed for Agentic network intelligence. By decoupling network knowledge acquisition and management from intelligence logic, KP-A streamlines knowledge development, reduces maintenance complexity for radio engineers, and enhances interoperability for the network intelligence agents. We build and demonstrate KP-A in 2 representative intelligence tasks: (1) live network knowledge Q&A, and (2) edge-AI service orchestration in RAN. All implementation artifacts have been open-sourced to support reproducibility and future standardization efforts.