MX-AI: Agentic Observability and Control Platform for Open and AI-RAN

2025-08-08

AI Summary

To address the lack of real-time autonomous management and control in 6G Open Radio Access Network (O-RAN), this paper proposes the first end-to-end AI-driven RAN agent system, pioneering the integration of large language models (LLMs) into the RAN Management and Orchestration (RAN-MO) layer. Built on a 5G testbed leveraging OpenAirInterface and FlexRIC, the system implements an LLM-based agent graph that enables natural-language intent input for resource observability and dynamic reconfiguration. It unifies the Service Management and Orchestration (SMO) architecture with a natural-language interface, endowing the system with expert-level reasoning and decision-making capabilities. Experimental evaluation demonstrates an average response quality of 4.1/5.0 across 50 real-world operational queries, 100% decision-to-action accuracy, and an end-to-end latency of only 8.8 seconds (using GPT-4.1), matching human expert performance.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Future 6G radio access networks (RANs) will be artificial intelligence (AI)-native: observed, reasoned about, and re-configured by autonomous agents cooperating across the cloud-edge continuum. We introduce MX-AI, the first end-to-end agentic system that (i) instruments a live 5G Open RAN testbed based on OpenAirInterface (OAI) and FlexRIC, (ii) deploys a graph of Large-Language-Model (LLM)-powered agents inside the Service Management and Orchestration (SMO) layer, and (iii) exposes both observability and control functions for 6G RAN resources through natural-language intents. On 50 realistic operational queries, MX-AI attains a mean answer quality of 4.1/5.0 and 100 % decision-action accuracy, while incurring only 8.8 seconds end-to-end latency when backed by GPT-4.1. Thus, it matches human-expert performance, validating its practicality in real settings. We publicly release the agent graph, prompts, and evaluation harness to accelerate open research on AI-native RANs. A live demo is presented here: https://www.youtube.com/watch?v=CEIya7988Ug&t=285s&ab_channel=BubbleRAN

Problem

• Develops an AI-native platform for 6G RAN observability and control
• Integrates LLM-powered agents in 5G Open RAN testbed management
• Achieves human-expert performance in decision-action accuracy and latency

Innovation

• End-to-end agentic system for 6G RAN
• LLM-powered agents in SMO layer
• Natural-language observability and control