An accurate AOA localization method based on unreliable sensor detection

Abstract

Least Squares (LS) triangulation is one of the most used methods for AOA (Angle of Arrival) localization. However, the localization results can be affected by bearing outliers. So it is important to identify whether there are unreliable bearing measurements, filter these out and only use the reliable sensor data to estimate sound source position. This paper proposes a new method based on the characteristics of estimated positions of different sensor combinations to identifying unreliable sensors. The target position is calculated by estimated positions obtained from reliable sensors. The simulation results for indoor environments show that the localization error can be reduced to less than 0.1 m in an empty shoebox model even if both the sensors and the sound source are all close to the wall.