An Indoor Localization Algorithm for Lighting Control using RFID

Abstract

Occupant's identity and location are important information for lighting control in order to reduce the energy consumption while increasing livelihood. While active RFID system provides occupant's identity, it is nontrivial to localize the occupant's location in an indoor environment due to the multipath effect, the changing environment, and the unreliable communication link. In this paper, we implement a system with multiple active RFID readers, and develop a localization algorithm based on support vector machine (SVM). The algorithm uses round-robin comparison to localize the occupant to one of the multiple regions in a floor. The geometric relationship among the rooms and the historical localization data are used to further improve the localization accuracy. Numerical results demonstrate a high localization accuracy of this algorithm. We hope this work sheds insight on lighting control for energy saving and an increased livelihood.