Enhancing WiFi Access Point Localization With AI-Based Filtering

Cheng Yu Yang, Wan Ting Shih, Chao Kai Wen*, Shang Ho Tsai, Chau Yuen

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Accurate location determination of WiFi access points (APs) is crucial for a variety of industrial and commercial applications. Although WiFi beacons are the most common signals emitted by APs, using them for AP position estimation is challenging due to limited bandwidth. This limitation leads to unreliable parameter extraction and hampers AP positioning accuracy. We propose an Artificial Intelligence (AI)-based filter designed to eliminate erroneously extracted parameters. This innovative filter can adapt its criteria based on previously known AP position information, facilitating intelligent collaboration with existing AP localization methods. It begins with a coarse filtering approach to quickly ascertain a rough AP position, then incrementally refines its criteria to enhance AP positioning precision, rigorously preventing the disqualification of data when the AP position is already at a high level of precision. Simulations and experiments consistently confirm that the proposed AI-based filter significantly improves AP positioning accuracy, achieving decimeter-level precision even with only WiFi beacons operating on a 20MHz bandwidth.

Original languageEnglish
Pages (from-to)1332-1336
Number of pages5
JournalIEEE Communications Letters
Volume28
Issue number6
DOIs
StatePublished - 1 Jun 2024

Keywords

  • WiFi
  • artificial intelligence (AI)
  • localization
  • simultaneous localization and mapping (SLAM)

Fingerprint

Dive into the research topics of 'Enhancing WiFi Access Point Localization With AI-Based Filtering'. Together they form a unique fingerprint.

Cite this