Highly Accurate Distance Estimation Using Spatial Filtering and GNSS in Urban Environments

Abstract

In this paper, an algorithm for accurate distance estimation of a vehicle to a road side unit (RSU) in urban multipath environments is presented. The algorithm is based on spatial filtering using a uniform circular array (UCA) antenna and a cross correlation based time delay detection. The time base for the cross correlation is built using Global Navigation Satellite System (GNSS) based positioning. Furthermore, an algorithm for non- line-of-sight (NLOS) detection based on a fingerprinting technique is presented to identify shadowing situations. For evaluation purposes, a simulation for a vehicle following a straightway maneuver within an urban orthogonal traffic intersection scenario is carried out. The B1 line- of-sight (LOS) specified in WINNER II is used for multipath channel modeling. Simulation results show that the obtained distance estimation accuracy of a vehicle is better than 50cm within a radius of 500m around the traffic intersection origin and the proposed NLOS detection algorithm exhibits a 0% false alarm probability (FAP) within this range.