Performance Comparison of Two Methods for Indoors Localization Using Wireless Communications

• Type: Article in International Conference Proceedings Book
• Date: 2011

Authors

Abílio Azenha (Author) FEUP; Luis Peneda (Author) FEUP; Adriano Carvalho (Author) FEUP

Conference proceedings International

Title: 18th Saint Petersburg International Conference on Integrated Navigation Systems

Pages: 306-313

18th Saint Petersburg International Conference on Integrated Navigation Systems

São Petersburgo, Federação da Rússia, 30 de Maio a 01 de Junho de 2011

Scientific classification

FOS: Engineering and technology > Electrical engineering, Electronic engineering, Information engineering

CORDIS: Technological sciences > Engineering > Control engineering

Other information

Abstract (EN): This paper describes indoors localization methods for object tracking in some industrial applications such as automatic warehouses. Localization accuracies of down to one meter can be achieved with these methodologies based on current paper experiments. The two considered methods of indoors localization are based on ZigBee networks, the first one, and on generic wireless networks, the second one. They are compared in terms of performance characteristics such as accuracy, repeatability and easy of deployment. They adopt wireless communications framework and are therefore a sub-product of such systems. So, only software is additionally required. This makes the localization systems modular and of low-cost characteristics. It is shown that the method which can be built with generic wireless networks can be a promising solution due to its easy of deployment, relatively good accuracy and good repeatability. In addition, some considerations on filtering of signal strength used in localization algorithms are presented in order to smooth signal fluctuations which in turn can improve systems performance. Experimental results are carried out which show the performance of these two methods.

Language: English

Type (Professor's evaluation): Scientific