University of Alcalá
|Type of Publication:||In Proceedings|
|Book title:||International Workshop on Perception in Robotics, IEEE Intelligent Vehicles Symposium|
Traditionally, WiFi has been used for indoors localization purposes due to its important advantages. First, it is already deployed and its coverage is quickly growing. Second, measuring the WiFi signal strength is free of charge for every WiFi network. Unfortunately, it also has some disadvantages: when extending WiFi-based localization systems to large environments with a high number of APs and positions their accuracy decreases. This has been previously solved by manually dividing the environment into sub-regions. In this paper, an automatic division of the environment using different clustering techniques is proposed. Then the final position is computed by classifying in two different levels (first for a zone, then for a position inside it) using a hierarchical approach. The final performance of the system was improved by dividing the environment in zones obtaining an overall increase in the accuracy of approximately 5%. This indicates that by dividing the environment the overall performance can be improved.
|[Bibtex] [RIS] [MODS]|