Sample-and-Score: Surrogate-Guided Coordinated Spatial Reuse in IEEE 802.11 Networks

Maksymilian Wojnar

cs.NIOct 27, 2026

Sample-and-Score: Surrogate-Guided Coordinated Spatial Reuse in IEEE 802.11 Networks

Authors: Maksymilian Wojnar

Organizations: AGH University of Krakow, Poland

Abstract

Coordinated spatial reuse (Co-SR) in IEEE 802.11bn increases area throughput through simultaneous transmissions, yet the combinatorial configuration space makes real-time optimization challenging. We propose a sample-and-score approach: draw random Co-SR configurations, score each with a surrogate model, and select the best. The surrogate is trained offline and deployed without online fine-tuning or sequential exploration, yet performs close to multi-armed bandit and heuristic baselines that need multiple interaction steps or extensive measurements. On unseen topologies with 2-16 access points, it recovers 52-77% of a perfect surrogate's throughput. A perfect sample-and-score approach reaches 95% of the upper bound throughput with 8192 candidates.

Figures & tables

Panel of Figure 1 — Fig. 1: Overview of the proposed surrogate-based Co-SR optimization pipeline: random configurations are encoded as graphs, scored by the surrogate, and the best ones are selected for execution.

Parameter	Value
Band	5 GHz
PHY	IEEE 802.11be
Channel width	80 MHz
Spatial streams	1, SISO
Path loss model	TGax w/ Nakagami fading
Nakagami-m parameter	1.5

TABLE I: Simulation parameters.

Panel of Figure 3 — Fig. 2: Throughput vs. number of candidates $N$ in the residential scenario ( $2\times 2$ apartments) with top-1 selection. Error bars denote 95% confidence intervals of the mean.

Panel of Figure 4 — Fig. 3: Co-SR throughput across open-space scenarios with 2–16 APs (top-4 selection, number of candidates $N$ ranging from 64 to 512).

N	Small (4 APs)		Large (16 APs)
N	Full	Split	Full	Split
4	1.5	1.5	1.9	1.9
8	1.7	1.7	2.7	2.7
16	2.0	2.0	6.0	6.0
32	2.8	3.9	18.5	12.1
64	6.2	7.8	66.2	24.1

TABLE II: Surrogate computation time vs. H-MAB convergence time [

\mathrm{ms}

]. Dashes indicate out-of-memory conditions under full-batch inference.

Wireless Local Area Networks

See all 20 papers

FM4WiFi: Flow Matching for Multi-AP Coordination in Dense Deployments of Beyond Wi-Fi 8 Networks
Aug 4, 2026Maksymilian Wojnar, Krzysztof Rusek, Katarzyna Kosek-Szott +1Wireless Communications Wireless Resource Allocation
AGH University of Krakow, Poland
CityDeploy-Bench: Benchmarking Physics-Grounded Spatial Set Planning for Multi-Transmitter Network Deployment
Oct 8, 2026Chenyang Yuan, Xiaoyuan ChengBenchmark Design Wireless Communications
University of Sheffield · University College London
Learning Coverage- and Power-Optimal Transmitter Placement from City Maps: A Comparative Study of Direct and Indirect Neural Approaches
Apr 23, 2026Çağkan YaparWireless Communications Multi-Objective Optimization

Aug 4, 2026J/K move · Enter open · S save

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

Maksymilian Wojnar, Krzysztof Rusek, Katarzyna Kosek-Szott, Szymon Szott

AGH University of Krakow, Poland

Wireless Communications Wireless Resource Allocation Wireless Local Area Networks

Open paper AskPDF arXiv

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.

Abstract

Wireless Local Area Networks

Explore similar work

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

CityDeploy-Bench: Benchmarking Physics-Grounded Spatial Set Planning for Multi-Transmitter Network Deployment

Learning Coverage- and Power-Optimal Transmitter Placement from City Maps: A Comparative Study of Direct and Indirect Neural Approaches