Indoor-localization system for smart phones

Abstract

In this paper we present an indoor-localization system using sound signals which are outside of the audible range with an error in the range of centimeters. The innovative distinctive feature of the proposed system is based on the high accuracy due to the acoustic run-time measurements. Therefore this approach is more precise than other indoor-localization systems based on Wi-Fi or Bluetooth. Furthermore, it is user-friendly as the user needs only his smartphone and no additional hardware. With its high precision the system allows entirely new applications in indoor localization. In particular, it has the technological potential to help blind people to navigate through public and administrative buildings without the need of assistance from others. In this novel approach we implement an unscented Kalman filter which proves to be capable of tracking not only the position of the target but also the time offset of the smartphone clock, reducing the required number of available receivers for localization. Besides, the precise synchronization of the receivers allows us to track the current position of the smartphone user with a median error of 9 cm and a standard deviation of 9 cm in a real-world scenario. In this way, the user is enabled to track his current location in a building with a very high accuracy.