EvoRIC: Reinforcement Learning Fine-Tuned LLM-empowered RAN Intelligent Control Toward Autonomous O-RAN
This work aims to improve the generalization and domain-specific decision-making capabilities of RAN intelligent controllers for O-RAN, which is an incremental improvement for network operators.
This paper introduces EvoRIC, a hierarchical framework for RAN intelligent control that addresses the limitations of traditional ML and general-purpose LLMs in wireless networks. It employs a reinforcement learning-based fine-tuning mechanism where an LLM acts as a PPO agent, iteratively updating its parameters based on interactions with the wireless environment to align semantic reasoning with network performance objectives.
Despite recent advances in applying artificial intelligence (AI) techniques to radio access network (RAN), critical challenges remain: traditional machine learning (ML) algorithms suffer from limited generalization across varying network topologies, whereas general-purpose large language models (LLMs) face high computational demands and lack domain-specific knowledge. To address these gaps, this article introduces the evolving RAN intelligent controller (RIC) (EvoRIC) framework, a hierarchical architecture that enables continuous evolution by leveraging a non-real-time RIC (non-RT RIC) for global model updates and a near-real-time RIC (near-RT RIC) for local execution, dynamically empowering LLMs with domain-specific decision-making capabilities. Within this framework, we employ a reinforcement learning-based fine-tuning (RLFT) mechanism where an LLM operates as an actor within a proximal policy optimization (PPO) agent. By leveraging the interaction tuples collected from the wireless environment, the LLM's parameters are iteratively updated to align semantic reasoning with rigorous network performance objectives. We evaluate the generalization and efficacy of the proposed EvoRIC framework within integrated access and backhaul (IAB) networks, and finally, discuss the open challenges and future directions of the EvoRIC framework toward realizing autonomous O-RAN.