Concept of image based Non-Line-of-Sight (NLOS) localization in multipath environments

Abstract

Current bidirectional localization schemes are able to locate a mobile device using Line-of-Sight (LOS) or Non-Line-of-Sight (NLOS) Time-of-Arrival (TOA) and Angle-of-Arrival (AOA) information measured at both the mobile device and reference device. This information is used to derive line of possible mobile device position (LPMD). The intersection points of LPMDs are used to estimate mobile position. However these algorithms do not work well in a dense multipath environment with high levels of TOA and AOA measurement noise. In addition, these techniques require at least two single bounce reflection paths to locate the mobile position. This paper explores the feasibility of using multiple image theory to obtain the image point of the NLOS single bounce multipath to perform NLOS localization and overcome the abovementioned limitations. Simulation results have shown that there are few variance shape of image point with various combinations of TOA and AOA variance noises. With the discovery of these variance shapes in the image point, it opened up new method and possibility to perform NLOS localization in a more effective and accurate way.