Home // International Journal On Advances in Networks and Services, volume 3, numbers 1 and 2, 2010 // View article
Authors:
Jürgen Kemper
Nicolaj Kirchhof
Markus Walter
Holger Linde
Keywords: Human Assisted, Calibration, Localization, Infrared, Thermal, Thermopile, Tracking, Auto-Calibration
Abstract:
Besides aspects like accuracy and cost, a simple setup process is essential for the adoption of location systems. However, existing systems often require a time-consuming calibration process to determine sensor node positions or orientations. In this paper we present a software aided approach for the calibration of an angulation-based indoor location system, which only demands from the user to walk through the room. The novelty of the presented approach is that a completely passive sensor technology is applied whereas the calculations are exclusively based on angular measurements. The node localization is realized without any prior knowledge of sensor positions, orientations or the location of the moving person. Nevertheless, it can be shown that an accurate localization of the sensor nodes is possible. The presented algorithm is based on the Newton-Raphson method for solving non-linear equation systems. In order to improve the calibration results, a preselection of the calibration measurements is processed that realizes the identification and non consideration of unreliable measurements. The accuracy of the approach, its convergence probability and its runtime are evaluated by several simulations. Furthermore, real-world tests, using a location system that exploits the thermal radiation of humans for localization, are carried out to verify the simulation results.
Pages: 82 to 91
Copyright: Copyright (c) to authors, 2010. Used with permission.
Publication date: September 5, 2010
Published in: journal
ISSN: 1942-2644