Indoor localisation robustness and performance improvement

  • Authors:
  • Yiming Ji

  • Affiliations:
  • University of South Carolina Beaufort, One University Blvd, Bluffton, SC, 29909, USA

  • Venue:
  • International Journal of Communication Networks and Distributed Systems
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper studies performance robustness of dynamic indoor localisation systems that use only received signal strength (RSS). We introduce a simple radio distortion method to capture signal perturbations under uncontrolled environments where signal values could be changed dynamically. With the method, we examine several location searching algorithms, including 'lateration' and 'multidimensional scaling (MDS)' for distance-based systems, as well as 'clustering', 'least mean square errors', 'closeness elimination' and 'second lowest mean square errors' for the database-based system using two different environments including one regular office building and one underground floor-plan, we observed that 'MDS' and 'clustering' presented stable and smaller localisation errors under most experimental settings. Of systems studied in this paper, strong experimental evidence also indicated that the signal-location map method, a database-based system, outperforms all distance-based systems in this paper, especially when environment is dynamic or when radio signal values are distorted intentionally.