Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Next-Generation Wi-Fi Networks with Generative AI: Design and Insights

Jingyu Wang, Xuming Fang, Dusit Niyato, Tie Liu

TL;DR

This work investigates how generative AI can address the design and performance challenges of next-generation Wi-Fi networks, where extensive PHY/MAC parameterization and new features like MLO complicate optimization. It proposes a retrieval-augmented LLM framework to assist scenario analysis and problem formulation, paired with a diffusion-model–based DRL solver (GDM–DRL) to optimize network parameters, and validates the approach in high-density deployments using NS-3. The key contributions include the RA-LLM optimization framework, a practical problem formulation for throughput maximization under dense conditions, and demonstration that GDM-based DRL can outperform traditional methods and baseline 802.11 configurations. The paper also discusses future directions in interference mitigation, privacy protection via federated learning, and multi-modal data fusion to further empower GAI-assisted Wi-Fi networks, emphasizing practical impact for scalable, intelligent WLAN design.

Abstract

Generative artificial intelligence (GAI), known for its powerful capabilities in image and text processing, also holds significant promise for the design and performance enhancement of future wireless networks. In this article, we explore the transformative potential of GAI in next-generation Wi-Fi networks, exploiting its advanced capabilities to address key challenges and improve overall network performance. We begin by reviewing the development of major Wi-Fi generations and illustrating the challenges that future Wi-Fi networks may encounter. We then introduce typical GAI models and detail their potential capabilities in Wi-Fi network optimization, performance enhancement, and other applications. Furthermore, we present a case study wherein we propose a retrieval-augmented LLM (RA-LLM)-enabled Wi-Fi design framework that aids in problem formulation, which is subsequently solved using a generative diffusion model (GDM)-based deep reinforcement learning (DRL) framework to optimize various network parameters. Numerical results demonstrate the effectiveness of our proposed algorithm in high-density deployment scenarios. Finally, we provide some potential future research directions for GAI-assisted Wi-Fi networks.

Next-Generation Wi-Fi Networks with Generative AI: Design and Insights

TL;DR

This work investigates how generative AI can address the design and performance challenges of next-generation Wi-Fi networks, where extensive PHY/MAC parameterization and new features like MLO complicate optimization. It proposes a retrieval-augmented LLM framework to assist scenario analysis and problem formulation, paired with a diffusion-model–based DRL solver (GDM–DRL) to optimize network parameters, and validates the approach in high-density deployments using NS-3. The key contributions include the RA-LLM optimization framework, a practical problem formulation for throughput maximization under dense conditions, and demonstration that GDM-based DRL can outperform traditional methods and baseline 802.11 configurations. The paper also discusses future directions in interference mitigation, privacy protection via federated learning, and multi-modal data fusion to further empower GAI-assisted Wi-Fi networks, emphasizing practical impact for scalable, intelligent WLAN design.

Abstract

Generative artificial intelligence (GAI), known for its powerful capabilities in image and text processing, also holds significant promise for the design and performance enhancement of future wireless networks. In this article, we explore the transformative potential of GAI in next-generation Wi-Fi networks, exploiting its advanced capabilities to address key challenges and improve overall network performance. We begin by reviewing the development of major Wi-Fi generations and illustrating the challenges that future Wi-Fi networks may encounter. We then introduce typical GAI models and detail their potential capabilities in Wi-Fi network optimization, performance enhancement, and other applications. Furthermore, we present a case study wherein we propose a retrieval-augmented LLM (RA-LLM)-enabled Wi-Fi design framework that aids in problem formulation, which is subsequently solved using a generative diffusion model (GDM)-based deep reinforcement learning (DRL) framework to optimize various network parameters. Numerical results demonstrate the effectiveness of our proposed algorithm in high-density deployment scenarios. Finally, we provide some potential future research directions for GAI-assisted Wi-Fi networks.
Paper Structure (17 sections, 4 figures, 1 table)

This paper contains 17 sections, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Overview of Current and Future Wi-Fi Networks
  3. Current Status of Wi-Fi Technology
  4. Future Challenges for Wi-Fi Networks
  5. Applications of GAI for Wi-Fi Networks
  6. GAI Models and Applications
  7. LLM-assisted Wi-Fi Networks Design
  8. GAI-assisted Wi-Fi Networks Performance Enhancement
  9. Roles of GAI in Other Wi-Fi-based Applications
  10. Case Study
  11. LLM-enabled Wi-Fi Optimization Framework
  12. Wi-Fi Network Performance Enhancement in High-density Deployment Scenarios
  13. Future Directions
  14. Interference Mitigation
  15. Privacy Protection
  16. ...and 2 more sections

Figures (4)

  • Figure 1: The features of the current Wi-Fi generations and the applications of Wi-Fi networks. Part A compare the main features of Wi-Fi 4 to Wi-Fi 7. Part B summarizes the primary Wi-Fi applications, encompassing communication, sensing, and edge computing.
  • Figure 2: The proposed framework for Wi-Fi network optimization.
  • Figure 3: The case to use the proposed framework to assist in formulating problem.
  • Figure 4: GDM assisted throughput maximization for high-density deployment scenarios. We compare the convergence speed of GDM and DDPG, followed by throughput comparisons across different baselines and numbers of stations.