Reimagining RAN Automation in 6G: An Agentic AI Framework with Hierarchical Online Decision Transformer

TL;DR

This paper introduces an agentic AI framework for 6G wireless networks, utilizing a Hierarchical Online Decision Transformer to coordinate diverse AI agents guided by natural language inputs, improving network performance and self-healing capabilities.

Contribution

It presents a novel hierarchical AI framework with a decision transformer and retrieval-augmented generation for dynamic network management and operator intent validation.

Findings

Improved throughput, reduced delay, and higher energy efficiency compared to baselines.

Achieved 88.5% accuracy in operator intent validation.

Recovered 90% of performance during interruptions through self-healing.

Abstract

In this paper, we propose an Agentic Artificial Intelligence (AI) framework for wireless networks. The framework coordinates a pool of AI agents guided by Natural Language (NL) inputs from a human operator. At its core, the super agent is powered by a Hierarchical Online Decision Transformer (H-ODT). It orchestrates three categories of agents: (i) inter-slice, intra-slice resource allocation agents, (ii) network application orchestration agents, and (iii) self-healing agents. The orchestration takes place with the help of an Agentic Retrieval-Augmented Generation (RAG) module that integrates knowledge from heterogeneous sources. In this proposed methodology, the super agent directly interfaces with operators and generates sequential policies to activate relevant agents. The proposed framework is evaluated against three state-of-the-art baselines, showing improved throughput, reduced network delay, and higher energy efficiency at both slice-level and system-wide performance metrics. Also, the proposed Agentic framework introduces a bi-level human operator intent validation methodology, both at the slice-level and Key Performance Indicator (KPI)-level using generative AI-based time series predictors. We could rule out performance-degrading operator intents with an accuracy of 88.5%. Lastly, while being interrupted by any performance-degrading events, the self-healing capability of Agentic AI in our framework automatically recovers 90% of its previous performance, avoiding quality-of-service drifts when there is no human involvement.