Crowdsourcing-Based Indoor Localization with Knowledge-Aided Fingerprint Transfer

Chenlu Xiang, Shunqing Zhang*, Shugong Xu, Guoqiang Mao

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

7 Citations (Scopus)

Abstract

Precise indoor localization is a key requirement for the fifth-generation (5G) and beyond wireless communication systems with applications. To that end, many high accuracy signal fingerprint-based localization algorithms have been proposed. Most of these algorithms, however, face the problem of performance degradation in indoor environments, when the propagation environment changes with time. In order to address this issue, the crowdsourcing approach has been recently adopted, where the fingerprint database is frequently updated via user reporting. These crowdsourcing techniques still require precise indoor floor plans and fail to provide satisfactory accuracy. In this paper, we propose a low-complexity self-calibrating indoor crowdsourcing localization system that combines historical fingerprints with frequently updated fingerprints for high precision user positioning. We present a multi-kernel transfer learning approach that exploits the inner relationship between the historical and updated channel state information (CSI). Our indoor laboratory experimental results using Nexus 5 smartphones at 2.4GHz with 20MHz bandwidth have shown the feasibility of the proposed approach to achieve about one-meter level accuracy with a reasonable fingerprint update overhead.

Original languageEnglish
Pages (from-to)4281-4293
Number of pages13
JournalIEEE Sensors Journal
Volume22
Issue number5
DOIs
Publication statusPublished - 1 Mar 2022
Externally publishedYes

Keywords

  • channel state information
  • crowdsourcing
  • fingerprint database
  • Indoor localization
  • transfer learning

Fingerprint

Dive into the research topics of 'Crowdsourcing-Based Indoor Localization with Knowledge-Aided Fingerprint Transfer'. Together they form a unique fingerprint.

Cite this