FM4WiFi: Flow Matching for Multi-AP Coordination in Dense Deployments of Beyond Wi-Fi 8 Networks

2026-08-04

AI Summary

This work addresses the limited scalability of existing Wi-Fi 8 multi-AP coordination mechanisms, which support only pairwise collaboration and struggle with high signaling overhead, slow convergence, and computational complexity in dense networks—particularly under coordinated spatial reuse (Co-SR). To overcome these challenges, we propose FM4WiFi, the first end-to-end framework that introduces flow-matching generative modeling to wireless coordination. FM4WiFi employs an autoencoder to compress network state and a surrogate rate predictor to evaluate configuration quality, enabling efficient generation of complete Co-SR configurations—including power and rate control—in a single inference pass. Requiring neither real-time simulation nor a digital twin, our method achieves sub-second inference even in dense deployments with over 30 APs, matching or surpassing state-of-the-art performance while significantly enhancing scalability and practical deployability.

Abstract

Wi-Fi networks are moving beyond random channel access toward tightly coordinated operation across access points (APs), a shift reflected in Wi-Fi 8's multi-AP coordination (MAPC). However, the current MAPC specification restricts cooperation to AP pairs, fundamentally limiting the gains achievable in dense deployments and calling for scalable, network-wide coordination in beyond Wi-Fi 8 systems. We target coordinated spatial reuse (Co-SR), where APs transmit concurrently at reduced power. Effective Co-SR demands joint selection and configuration of AP-station transmissions, yet existing approaches simply do not scale: they rely on heavy signaling, slow convergence, unrealistic assumptions, and often require computation time that explodes with network size. We introduce FM4WiFi, a generative ML pipeline that addresses these limitations by producing high-quality Co-SR configurations in a single inference step. FM4WiFi integrates (i) an autoencoder that learns compact latent representations of network states, (ii) a flow-matching generative model that synthesizes feasible Co-SR configurations (including rate control, absent from prior work), and (iii) a surrogate rate predictor that allows rapid, large-scale Co-SR candidate evaluation without dependence on a live system or digital twin. Across extensive evaluations (including experimental validation), FM4WiFi matches or exceeds state-of-the-art baselines at medium-to-large scales and scales to 30+ APs with sub-second inference. Extensive ablation studies validate each modeling and optimization choice.

Problem

• Multi-AP Coordination
• Coordinated Spatial Reuse
• Dense Wi-Fi Networks
• Beyond Wi-Fi 8
• Scalability

Innovation

• Flow Matching
• Multi-AP Coordination
• Coordinated Spatial Reuse
• Generative Machine Learning
• Beyond Wi-Fi 8