Improving the Spatial Reuse in IEEE 802.11ax WLANs

Bardou, Anthony; Begin, Thomas; Busson, Anthony

doi:10.1145/3479239.3485715

Cited by 14 publications

(28 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While [21] studied the problem by using selfish rewards in a competitive environment, [22] considered shared rewards for the sake of maximizing fairness. Other RL-based approaches can be found in [23] and [24].…”

Section: Parametrized Spatial Reuse (Psr)mentioning

confidence: 99%

Federated Spatial Reuse Optimization in Next-Generation Decentralized IEEE 802.11 WLANs

Wilhelmi¹,

Hribar²,

Yılmaz³

et al. 2022

Preprint

View full text Add to dashboard Cite

As wireless standards evolve, more complex functionalities are introduced to address the increasing requirements in terms of throughput, latency, security, and efficiency. To unleash the potential of such new features, artificial intelligence (AI) and machine learning (ML) are currently being exploited for deriving models and protocols from data, rather than by hand-programming. In this paper, we explore the feasibility of applying ML in next-generation wireless local area networks (WLANs). More specifically, we focus on the IEEE 802.11ax spatial reuse (SR) problem and predict its performance through federated learning (FL) models. The set of FL solutions overviewed in this work is part of the 2021 International Telecommunication Union (ITU) AI for 5G Challenge.

show abstract

Section: Parametrized Spatial Reuse (Psr)mentioning

confidence: 99%

Federated Spatial Reuse Optimization in Next-Generation Decentralized IEEE 802.11 WLANs

Wilhelmi¹,

Hribar²,

Yılmaz³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However, given the vast diversity of WLAN topologies, the offline learning of the ANN appears as a clear limitation to the generalization of this strategy. An online learning procedure is proposed by [5], which uses reinforcement learning and more precisely the Multi-Armed Bandit (MAB) framework to find the optimal configuration of TX PWR and OBSS PD in a WLAN. The approach comprises two agents with one sampling promising configurations through a multivariate normal distribution, and the other identifying the best configuration among those already sampled with Thompson sampling and Normal-Gamma priors.…”

Section: Related Workmentioning

confidence: 99%

“…The approach comprises two agents with one sampling promising configurations through a multivariate normal distribution, and the other identifying the best configuration among those already sampled with Thompson sampling and Normal-Gamma priors. These two ML solutions [4,5] were tested on the network simulator ns-3 and lead to significant WLAN improvements. However, in order to perform their optimization, they both assume the presence of a central controller that has access and control over all the APs in the WLANs.…”

Section: Related Workmentioning

confidence: 99%

“…The push area designates the area which each AP can influence typically through the prescription of parameter configurations. In the case of centralized strategies (e.g., [4,5]), the pull and push areas naturally cover all the WLANs. We distinguish partially distributed strategies (e.g., [21]) wherein either the pull or push area extends all the WLANs with fully distributed strategies (e.g., [3,16,23]) in which both the pull and push areas differ from all the WLANs.…”

Section: Related Workmentioning

confidence: 99%

“…Additionally, most strategies were evaluated on relatively simple scenarios (with a few APs and a limited number of radio channels), often using a self-made network simulator. [3] STA STA Up Self-made 100/3 WCNC'21 [16] AP AP Down ns-3 6/1 Globecom'20 [4] WLAN WLAN Up/Down ns-3 3/1 ADHOC'19 [23] AP AP Down Self-made 8/1 JNCA'19 [21] WLAN AP Down Self-made 8/1 MSWiM'21 [5] WLAN WLAN Down ns-3 10/1…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

INSPIRE: Distributed Bayesian Optimization for ImproviNg SPatIal REuse in Dense WLANs

Bardou¹,

Begin²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

WLANs, which have overtaken wired networks to become the primary means of connecting devices to the Internet, are prone to performance issues due to the scarcity of space in the radio spectrum. As a response, IEEE 802.11ax and subsequent amendments aim at increasing the spatial reuse of a radio channel by allowing the dynamic update of two key parameters in wireless transmission: the transmission power (TX POWER) and the sensitivity threshold (OBSS PD).In this paper, we present INSPIRE, a distributed solution performing local Bayesian optimizations based on Gaussian processes to improve the spatial reuse in WLANs. INSPIRE makes no explicit assumptions about the topology of WLANs and favors altruistic behaviors of the access points, leading them to find adequate configurations of their TX POWER and OBSS PD parameters for the "greater good" of the WLANs. We demonstrate the superiority of INSPIRE over other state-of-theart strategies using the ns-3 simulator and two examples inspired by real-life deployments of dense WLANs. Our results show that, in only a few seconds, INSPIRE is able to drastically increase the quality of service of operational WLANs by improving their fairness and throughput.

show abstract