Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Generative Intent Prediction Agentic AI empowered Edge Service Function Chain Orchestration

Yan Sun, Shaoyong Guo, Sai Huang, Zhiyong Feng, Feng Qi, Xuesong Qiu

TL;DR

The paper addresses the challenge of managing edge services under high user mobility and pervasive implicit intents by proposing a Generative Intent Prediction Agent (GIPA). GIPA constructs a multidimensional intent space and uses a Generative Diffusion Model (GDM) to infer implicit intents from rich context, then embeds these predictions as global prompts to steer proactive SFC orchestration. The Generative Predictive SFC Orchestration (GPSO) framework combines a GIPM for implicit-intent prediction with a PSOM that deploys VNFs using PPO, guided by a unified reward that emphasizes delay minimization and service reliability. Experimental results in a simulated edge environment show that GIPA outperforms baselines in both execution delay and service success rate, especially under high concurrency and mobility, demonstrating the practical value of proactive, intent-aware edge management. The approach advances edge intelligence by linking semantic intent understanding with predictive orchestration to maintain service continuity in dynamic networks.

Abstract

With the development of artificial intelligence (AI), Agentic AI (AAI) based on large language models (LLMs) is gradually being applied to network management. However, in edge network environments, high user mobility and implicit service intents pose significant challenges to the passive and reactive management of traditional AAI. To address the limitations of existing approaches in handling dynamic demands and predicting users' implicit intents, in this paper we propose an edge service function chain (SFC) orchestration framework empowered by a Generative Intent Prediction Agent (GIPA). Our GIPA aims to shift the paradigm from passive execution to proactive prediction and orchestration. First, we construct a multidimensional intent space that includes functional preferences, QoS sensitivity, and resource requirements, enabling the mapping from unstructured natural language to quantifiable physical resource demands. Second, to cope with the complexity and randomness of intent sequences, we design an intent prediction model based on a Generative Diffusion Model (GDM), which reconstructs users' implicit intents from multidimensional context through a reverse denoising process. Finally, the predicted implicit intents are embedded as global prompts into the SFC orchestration model to guide the network in proactively and ahead-of-time optimizing SFC deployment strategies. Experiment results show that GIPA outperforms existing baseline methods in highly concurrent and highly dynamic scenarios.

Generative Intent Prediction Agentic AI empowered Edge Service Function Chain Orchestration

TL;DR

The paper addresses the challenge of managing edge services under high user mobility and pervasive implicit intents by proposing a Generative Intent Prediction Agent (GIPA). GIPA constructs a multidimensional intent space and uses a Generative Diffusion Model (GDM) to infer implicit intents from rich context, then embeds these predictions as global prompts to steer proactive SFC orchestration. The Generative Predictive SFC Orchestration (GPSO) framework combines a GIPM for implicit-intent prediction with a PSOM that deploys VNFs using PPO, guided by a unified reward that emphasizes delay minimization and service reliability. Experimental results in a simulated edge environment show that GIPA outperforms baselines in both execution delay and service success rate, especially under high concurrency and mobility, demonstrating the practical value of proactive, intent-aware edge management. The approach advances edge intelligence by linking semantic intent understanding with predictive orchestration to maintain service continuity in dynamic networks.

Abstract

With the development of artificial intelligence (AI), Agentic AI (AAI) based on large language models (LLMs) is gradually being applied to network management. However, in edge network environments, high user mobility and implicit service intents pose significant challenges to the passive and reactive management of traditional AAI. To address the limitations of existing approaches in handling dynamic demands and predicting users' implicit intents, in this paper we propose an edge service function chain (SFC) orchestration framework empowered by a Generative Intent Prediction Agent (GIPA). Our GIPA aims to shift the paradigm from passive execution to proactive prediction and orchestration. First, we construct a multidimensional intent space that includes functional preferences, QoS sensitivity, and resource requirements, enabling the mapping from unstructured natural language to quantifiable physical resource demands. Second, to cope with the complexity and randomness of intent sequences, we design an intent prediction model based on a Generative Diffusion Model (GDM), which reconstructs users' implicit intents from multidimensional context through a reverse denoising process. Finally, the predicted implicit intents are embedded as global prompts into the SFC orchestration model to guide the network in proactively and ahead-of-time optimizing SFC deployment strategies. Experiment results show that GIPA outperforms existing baseline methods in highly concurrent and highly dynamic scenarios.
Paper Structure (22 sections, 36 equations, 6 figures, 1 table, 2 algorithms)

This paper contains 22 sections, 36 equations, 6 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Works
  3. Agentic AI
  4. Intent-Driven Network
  5. Service Function Chain
  6. Problem Formulation
  7. Overview
  8. Intent Prediction Modeling
  9. Intent Space Construction
  10. GIPA Selection
  11. QoS Modeling
  12. Computation Delay
  13. Communication Delay
  14. Generative Predictive SFC Orchestration Model
  15. Generative Intent Prediction Model
  16. ...and 7 more sections

Figures (6)

  • Figure 1: Overview of GIPA empowered edge service function chain orchestration framework.
  • Figure 2: Overview of Generative Predictive SFC Orchestration Model.
  • Figure 3: Performance comparison between different methods. (a) Execution delay comparison. (b) Success rate comparison.
  • Figure 4: Performance of several methods when orchestrating navigation services.
  • Figure 5: Performance of several methods when orchestrating navigation services.
  • ...and 1 more figures