Locating smartphone users will enable numerous potential applications such as monitoring customers in shopping malls. However, conventional received signal strength (RSS)-based room-level localization methods are not likely to distinguish neighboring zones accurately due to similar RSS fingerprints. We solve this problem by proposing a system called feature-based room-level localization (FRL). FRL is based on an observation that different rooms vary in internal structures and human activities which can be reflected by RSS fluctuation ranges and user dwell time respectively. These two features combing with RSS can be exploited to improve the localization accuracy. To enable localization of unmodified smartphones, FRL utilizes probe requests, which are periodically broadcast by smartphones to discover nearby access points (APs). Experiments indicate that FRL can reliably locate users in neighboring zones and achieve a 10% accuracy gain, compared with conventional methods like the histogram method.
|Name||Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST|
|Conference||International Conference on Sustainable Solutions Beyond Mobility of Goods, SustainableMoG 2015|
|Period||13/10/15 → 14/10/15|
- Room-level localization
- Computer Networks and Communications