An Accurate Indoor-Localization Scheme with NLOS Detection and Elimination Exploiting Stochastic Characteristics

Manato HORIBA  Eiji OKAMOTO  Toshiko SHINOHARA  Katsuhiko MATSUMURA  

Publication
IEICE TRANSACTIONS on Communications   Vol.E98-B   No.9   pp.1758-1767
Publication Date: 2015/09/01
Online ISSN: 1745-1345
DOI: 10.1587/transcom.E98.B.1758
Type of Manuscript: Special Section PAPER (Special Section on Emerging Technologies on Ambient Sensor Networks toward Future Generation)
Category: 
Keyword: 
sensor network,  indoor localization,  NLOS environment,  iterative minimum residual scheme,  stochastic characteristic,  

Full Text: PDF(2.8MB)>>
Buy this Article




Summary: 
In indoor localization using sensor networks, performance improvements are required for non-line-of-sight (NLOS) environments in which the estimation error is high. NLOS mitigation schemes involve the detection and elimination of the NLOS measurements. The iterative minimum residual (IMR) scheme, which is often applied to the localization scheme using the time of arrival (TOA), is commonly employed for this purpose. The IMR scheme is a low-complexity scheme and its NLOS detection performance is relatively high. However, when there are many NLOS nodes in a sensor field, the NLOS detection error of the IMR scheme increases and the estimation accuracy deteriorates. Therefore, we propose a new scheme that exploits coarse NLOS detection based on stochastic characteristics prior to the application of the IMR scheme to improve the localization accuracy. Improved performances were confirmed in two NLOS channel models by performing numerical simulations.