Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

IntAgent: NWDAF-Based Intent LLM Agent Towards Advanced Next Generation Networks

Abdelrahman Soliman, Ahmed Refaey, Aiman Erbad, Amr Mohamed

TL;DR

The paper tackles autonomous intent-based networking in next-generation networks by integrating a large language model (LLM) driven agent with the 3GPP Network Data Analytics Function (NWDAF). It introduces IntAgent, an intelligent agent that embeds an intent tools engine inside the NWDAF analytics framework and communicates via the Model Context Protocol (MCP) to access live analytics and scheduling/monitoring tools. The authors validate the approach with two use cases—ML-based traffic prediction and scheduled policy enforcement—on a Kubernetes/Open5GS testbed with 10 UEs, demonstrating end-to-end autonomous fulfillment of network intents and real-time policy orchestration. Limitations include LLM hallucinations and planning overhead; future work explores domain-specific fine-tuning, cross-model comparisons, and expanding IntAgent to a collaborative multi-agent system for cross-domain intents.

Abstract

Intent-based networks (IBNs) are gaining prominence as an innovative technology that automates network operations through high-level request statements, defining what the network should achieve. In this work, we introduce IntAgent, an intelligent intent LLM agent that integrates NWDAF analytics and tools to fulfill the network operator's intents. Unlike previous approaches, we develop an intent tools engine directly within the NWDAF analytics engine, allowing our agent to utilize live network analytics to inform its reasoning and tool selection. We offer an enriched, 3GPP-compliant data source that enhances the dynamic, context-aware fulfillment of network operator goals, along with an MCP tools server for scheduling, monitoring, and analytics tools. We demonstrate the efficacy of our framework through two practical use cases: ML-based traffic prediction and scheduled policy enforcement, which validate IntAgent's ability to autonomously fulfill complex network intents.

IntAgent: NWDAF-Based Intent LLM Agent Towards Advanced Next Generation Networks

TL;DR

The paper tackles autonomous intent-based networking in next-generation networks by integrating a large language model (LLM) driven agent with the 3GPP Network Data Analytics Function (NWDAF). It introduces IntAgent, an intelligent agent that embeds an intent tools engine inside the NWDAF analytics framework and communicates via the Model Context Protocol (MCP) to access live analytics and scheduling/monitoring tools. The authors validate the approach with two use cases—ML-based traffic prediction and scheduled policy enforcement—on a Kubernetes/Open5GS testbed with 10 UEs, demonstrating end-to-end autonomous fulfillment of network intents and real-time policy orchestration. Limitations include LLM hallucinations and planning overhead; future work explores domain-specific fine-tuning, cross-model comparisons, and expanding IntAgent to a collaborative multi-agent system for cross-domain intents.

Abstract

Intent-based networks (IBNs) are gaining prominence as an innovative technology that automates network operations through high-level request statements, defining what the network should achieve. In this work, we introduce IntAgent, an intelligent intent LLM agent that integrates NWDAF analytics and tools to fulfill the network operator's intents. Unlike previous approaches, we develop an intent tools engine directly within the NWDAF analytics engine, allowing our agent to utilize live network analytics to inform its reasoning and tool selection. We offer an enriched, 3GPP-compliant data source that enhances the dynamic, context-aware fulfillment of network operator goals, along with an MCP tools server for scheduling, monitoring, and analytics tools. We demonstrate the efficacy of our framework through two practical use cases: ML-based traffic prediction and scheduled policy enforcement, which validate IntAgent's ability to autonomously fulfill complex network intents.
Paper Structure (17 sections, 5 figures, 1 algorithm)

This paper contains 17 sections, 5 figures, 1 algorithm.

Table of Contents

  1. INTRODUCTION
  2. SYSTEM ARCHITECTURE
  3. Overview of the Architecture
  4. NWDAF Components
  5. Intent Tools Engine
  6. PROPOSED AGENT
  7. High-Level Architecture
  8. Core Components of IntAgent
  9. LLM prompt
  10. Tools Set
  11. Agentic Loop
  12. EXPERIMENTS AND ANALYSIS
  13. Deployment setup
  14. Use case 1
  15. Use case 2
  16. ...and 2 more sections

Figures (5)

  • Figure 1: System model showing IntAgent integrated with NWDAF to enable intent-driven management through tools, event subscription, and data collection.
  • Figure 2: IntAgent architecture.
  • Figure 3: IntAgent tools connection schema.
  • Figure 4: Plot obtained from IntAgent final answer for Intent 1.
  • Figure 5: Policy change for scheduled data rate increase in streaming slice.