Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Learning Multi-Access Point Coordination in Agentic AI Wi-Fi with Large Language Models

Yifan Fan, Le Liang, Peng Liu, Xiao Li, Ziyang Guo, Qiao Lan, Shi Jin, Wen Tong

TL;DR

This work addresses adaptive MAPC in dense OBSS Wi-Fi deployments where static protocols fail under dynamic interference. It introduces a multi-LLM-agent framework in which each AP hosts an autonomous LLM with memory and tool use that negotiates transmission schedules through natural-language dialogue. Each negotiation round operates over $L$ slots, with a binary action vector $\mathbf{a}_k=[a_{k,1},\dots,a_{k,L}] \in \{0,1\}^L$ and dynamic switching between Co-TDMA and Co-SR based on perceived interference. Simulation results show superior throughput compared with optimized Wi-Fi 6 OBSS/PD baselines and demonstrate robust coexistence with legacy CSMA/CA APs. The findings validate a scalable, intelligent approach to decentralized wireless coordination with potential impact on future Wi‑Fi networks.

Abstract

Multi-access point coordination (MAPC) is a key technology for enhancing throughput in next-generation Wi-Fi within dense overlapping basic service sets. However, existing MAPC protocols rely on static, protocol-defined rules, which limits their ability to adapt to dynamic network conditions such as varying interference levels and topologies. To address this limitation, we propose a novel Agentic AI Wi-Fi framework where each access point, modeled as an autonomous large language model agent, collaboratively reasons about the network state and negotiates adaptive coordination strategies in real time. This dynamic collaboration is achieved through a cognitive workflow that enables the agents to engage in natural language dialogue, leveraging integrated memory, reflection, and tool use to ground their decisions in past experience and environmental feedback. Comprehensive simulation results demonstrate that our agentic framework successfully learns to adapt to diverse and dynamic network environments, significantly outperforming the state-of-the-art spatial reuse baseline and validating its potential as a robust and intelligent solution for future wireless networks.

Learning Multi-Access Point Coordination in Agentic AI Wi-Fi with Large Language Models

TL;DR

This work addresses adaptive MAPC in dense OBSS Wi-Fi deployments where static protocols fail under dynamic interference. It introduces a multi-LLM-agent framework in which each AP hosts an autonomous LLM with memory and tool use that negotiates transmission schedules through natural-language dialogue. Each negotiation round operates over slots, with a binary action vector and dynamic switching between Co-TDMA and Co-SR based on perceived interference. Simulation results show superior throughput compared with optimized Wi-Fi 6 OBSS/PD baselines and demonstrate robust coexistence with legacy CSMA/CA APs. The findings validate a scalable, intelligent approach to decentralized wireless coordination with potential impact on future Wi‑Fi networks.

Abstract

Multi-access point coordination (MAPC) is a key technology for enhancing throughput in next-generation Wi-Fi within dense overlapping basic service sets. However, existing MAPC protocols rely on static, protocol-defined rules, which limits their ability to adapt to dynamic network conditions such as varying interference levels and topologies. To address this limitation, we propose a novel Agentic AI Wi-Fi framework where each access point, modeled as an autonomous large language model agent, collaboratively reasons about the network state and negotiates adaptive coordination strategies in real time. This dynamic collaboration is achieved through a cognitive workflow that enables the agents to engage in natural language dialogue, leveraging integrated memory, reflection, and tool use to ground their decisions in past experience and environmental feedback. Comprehensive simulation results demonstrate that our agentic framework successfully learns to adapt to diverse and dynamic network environments, significantly outperforming the state-of-the-art spatial reuse baseline and validating its potential as a robust and intelligent solution for future wireless networks.

Paper Structure

This paper contains 14 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. System Model
  3. Agentic Workflow for Multi-AP Coordination
  4. The Multi-Agent Coordination Protocol
  5. LLM Agent Cognitive Architecture
  6. The Brain and its Reasoning Workflow
  7. Memory Module
  8. Tool Use Module
  9. Prompt Design
  10. Emergent Coordination Behaviors
  11. Simulation Results
  12. Simulation Setup
  13. Simulation Results
  14. Conclusion

Figures (3)

  • Figure 1: Agentic workflow and coordination protocol for multi-AP cooperation.
  • Figure 2: Transmission strategies of the agentic MAPC protocol under the Co-TDMA-favored and Co-SR-favored scenarios. For each negotiation round, the two columns within the bold frame display the transmission states in the 5 slots of the sharing AP (left) and the shared AP (right), respectively.
  • Figure 3: Coexistence between LLM-driven APs and legacy CSMA/CA APs in the Co-TDMA-favored scenario.