Acoustic indoor-localization system for smart phones

Abstract

In this paper we present our acoustic indoor-localization system ASSIST (Acoustic Self-calibrating System for Indoor Smart phone Tracking) running on embedded ARM CPUs. The developed system uses acoustic signals beyond the audible range to localize COTS (commercial off-the-shelf) smart phones. The sound receivers installed with the infrastructure are connected using a Wi-Fi network. Furthermore, the sound receivers synchronize the clocks and exchange the time differences of arrival of the received sound signals from the Wi-Fi network. In this way, using an iterative multilateration algorithm, the locations of smart phones are calculated. We present our developed novel receiver hardware. The integrated single-board computer based on ARM replaces the external computer units to synchronize the receivers with the network and calculates the position via a TDoA (Time Difference of Arrival) algorithm. By powering the receivers via power-over-Ethernet the installation effort is minimized. The integrated inertial sensors of the smart phone can be used to calculate the position in a non-line of sight condition. Therefore different methods are available to support the position calculation for ASSIST.