Open AccessVisible light positioning based on architecture information: method and performance
Author(s) -
Seguel Fabian,
Krommenacker Nicolas,
Charpentier Patrick,
Soto Ismael
Publication year - 2019
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2018.5623
Subject(s) - trilateration , computer science , centroid , robustness (evolution) , detector , ranging , algorithm , computer vision , position (finance) , range (aeronautics) , signal strength , field of view , artificial intelligence , real time computing , acoustics , physics , wireless , telecommunications , biochemistry , chemistry , materials science , finance , node (physics) , economics , composite material , gene
In this paper a novel 2‐D range free positioning method which uses visible light communications network architecture is presented. The algorithm performance for lattice and hexagonal LED lighting infrastructure deployment is obtained by computational simulation. The robustness of our method against changes on receiver's field of view and mismatch between the assumed vertical distance and real vertical position is simulated in a room of dimensions 5 × 5 × 2.5 m. A localization interest zone with receiver's height between 0.5 and 1.1 m measured from the floor is considered as variation on vertical distance. Moreover, FOV's ranging from 90° to 130° are used to measure the behaviour of the method when different built‐in photo detectors are incorporated in the mobile device. Our algorithm overcomes three traditional localization methods, i.e., convex position estimation, centroid and received signal strength based trilateration when changes in the FOV occurs. In addition to this, it shows better accuracy than other range free algorithms when an assumed height difference mismatch of up to 30 cm exist. The method shows better coverage than range based methods being able to reduce border effects and perform localization with reliable accuracy using only one connected cell and a single photo detector.