A novel approach to NLOS identification in sensor localization

Sensor localization is a basic and important task of wireless sensor networks, and abundant localization algorithms have been proposed based on various ranging techniques, including time-of-arrival (TOA), time-difference-of-arrival (TDOA), received signal strength (RSS), angle-of-arrival (AOA) and etc. The accuracy of these ranging techniques rely on the line-of-sight (LOS) propagation of signals, while in practical environments, signals may be propagated in non-line-of-sight (NLOS) channels, with the result that both ranging and localization accuracy can be seriously degraded. In this paper, we consider TOA-based sensor localization and propose a novel NLOS identification method using Cayley-Menger determinant (CMD). To be specific, the errors induced by NLOS in distance measurements are positive and much larger than errors from TOA systems, and as a result, the values of corresponding CMD behave differently under LOS/NLOS channels. Hence, we leverage this character to establish a statistical hypothesis testing model to identify NLOS channels. Finally, a simulation analysis is conducted to verify the effectiveness of the proposed method.