Representation Learning with Attention for Spatial Reuse Optimization in Dense WLANs

Stephen Azeez, Shagufta Henna

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

IEEE802.11ax is designed to support self-configuration and adaptation functionality in dense deployment to enhance dynamic network conditions. Presence of variable transmission range is one of the major bottlenecks to network performance. In the absence of proper power management, co-existing IEEE 802.11ax access points cause co-channel interference, degrading throughput. Therefore, it is essential to consider the impact of the variable transmit power of neighboring nodes to optimize the network performance. This work proposes an affinityGNN-attention mechanism to capture neighborhood transmit power to generate an expressive network representation. Experiments results show that the attention module integration improves the prediction accuracy and robustness of the baseline affinityGNN model.

Original languageEnglish
Title of host publicationProceedings of 8th International Congress on Information and Communication Technology - ICICT 2023
EditorsXin-She Yang, R. Simon Sherratt, Nilanjan Dey, Amit Joshi
PublisherSpringer Science and Business Media Deutschland GmbH
Pages949-959
Number of pages11
ISBN (Print)9789819930906
DOIs
Publication statusPublished - 2023
Event8th International Congress on Information and Communication Technology, ICICT 2023 - London, United Kingdom
Duration: 20 Feb 202323 Feb 2023

Publication series

NameLecture Notes in Networks and Systems
Volume694 LNNS
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

Conference8th International Congress on Information and Communication Technology, ICICT 2023
Country/TerritoryUnited Kingdom
CityLondon
Period20/02/2323/02/23

Keywords

  • Coexistence interference
  • IEEE802.11
  • Network management

Fingerprint

Dive into the research topics of 'Representation Learning with Attention for Spatial Reuse Optimization in Dense WLANs'. Together they form a unique fingerprint.

Cite this