Acoustic Signal Positioning and Calibration with IMU in NLOS Environment

Abstract

Acoustic signal has become a research hot pursuit of Indoor Location Based Service (ILBS) ascribed to its low synchronization rate, excellent performance and low cost, especially in large-scaled complicated indoor environments. Nevertheless, sound can easily be blocked and absorbed, producing non line-of-sight (NLOS) phenomenon. It brings great challenges to acoustic indoor positioning technology. In order to deal with NLOS, we introduce an Inertial Measurement Unit (TMU) to calibrate the huge error caused by shelters. Estimating the approximate result under NLOS based on the pedestrian posture calculated by adopting pedestrian dead reckoning (PDR) algorithm is a good choice. The coordinate is fed back to acoustic system to obtain the accurate position information. According to numerous of experiments, we achieve an accuracy of approximately 30 centimeters with 95% probability in NLOS positioning error within 50 meters from the anchor point (AP).