Advanced search
1 file | 1.94 MB

An unsupervised learning technique to optimize radio maps for indoor localization

Jens Trogh (UGent) , Wout Joseph (UGent) , Luc Martens (UGent) and David Plets (UGent)
(2019) SENSORS. 19(4).
Author
Organization
Abstract
A major burden of signal strength-based fingerprinting for indoor positioning is the generation and maintenance of a radio map, also known as a fingerprint database. Model-based radio maps are generated much faster than measurement-based radio maps but are generally not accurate enough. This work proposes a method to automatically construct and optimize a model-based radio map. The method is based on unsupervised learning, where random walks, for which the ground truth locations are unknown, serve as input for the optimization, along with a floor plan and a location tracking algorithm. No measurement campaign or site survey, which are labor-intensive and time-consuming, or inertial sensor measurements, which are often not available and consume additional power, are needed for this approach. Experiments in a large office building, covering over 1100 m(2), resulted in median accuracies of up to 2.07 m, or a relative improvement of 28.6% with only 15 min of unlabeled training data.
Keywords
TRACKING, FILTER, localization, radio map, unsupervised learning, tracking, positioning, rss, fingerprinting, indoor environment

Downloads

  • WICA 833.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 1.94 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Trogh, Jens, Wout Joseph, Luc Martens, and David Plets. 2019. β€œAn Unsupervised Learning Technique to Optimize Radio Maps for Indoor Localization.” Sensors 19 (4).
APA
Trogh, J., Joseph, W., Martens, L., & Plets, D. (2019). An unsupervised learning technique to optimize radio maps for indoor localization. SENSORS, 19(4).
Vancouver
1.
Trogh J, Joseph W, Martens L, Plets D. An unsupervised learning technique to optimize radio maps for indoor localization. SENSORS. Basel: Mdpi; 2019;19(4).
MLA
Trogh, Jens et al. β€œAn Unsupervised Learning Technique to Optimize Radio Maps for Indoor Localization.” SENSORS 19.4 (2019): n. pag. Print.
@article{8611127,
  abstract     = {A major burden of signal strength-based fingerprinting for indoor positioning is the generation and maintenance of a radio map, also known as a fingerprint database. Model-based radio maps are generated much faster than measurement-based radio maps but are generally not accurate enough. This work proposes a method to automatically construct and optimize a model-based radio map. The method is based on unsupervised learning, where random walks, for which the ground truth locations are unknown, serve as input for the optimization, along with a floor plan and a location tracking algorithm. No measurement campaign or site survey, which are labor-intensive and time-consuming, or inertial sensor measurements, which are often not available and consume additional power, are needed for this approach. Experiments in a large office building, covering over 1100 m(2), resulted in median accuracies of up to 2.07 m, or a relative improvement of 28.6\% with only 15 min of unlabeled training data.},
  articleno    = {752},
  author       = {Trogh, Jens and Joseph, Wout and Martens, Luc and Plets, David},
  issn         = {1424-8220},
  journal      = {SENSORS},
  language     = {eng},
  number       = {4},
  pages        = {17},
  publisher    = {Mdpi},
  title        = {An unsupervised learning technique to optimize radio maps for indoor localization},
  url          = {http://dx.doi.org/10.3390/s19040752},
  volume       = {19},
  year         = {2019},
}

Altmetric
View in Altmetric
Web of Science
Times cited: