Improving High Sensitivity Receiver Performance in Multipath Environment for Vehicular Applications

Abstract

Generally, standalone GNSS receivers architectures cannot provide a position accuracy suitable for use in vehicular applications in urban canyon scenarios. By using a block processing scheme, this paper aims to obtain better position performance in the urban canyon areas for the high sensitivity receivers. Specifically, the signal peaks (i.e., LOS peaks, multipath peaks, and cross-correlation peaks) are separated in the Doppler domain. Generally, the dominant peak is utilized in the high sensitivity receivers, however, this assumption is not always valid in the multipath environment. In particular, it is not assured that the dominant peak is the LOS peak. In this regard, the most accurate peaks (i.e., LOS peaks) are identified in this paper which yields better position performance compared to the dominant peak scheme. Least-squares method and Kalman filter are compared in this work by processing live data collected in an urban canyon. For the least-squares solution, the position performance is better than 40 m after using the most accurate peak, and for the Kalman filter solution, it is better than 10 m. GNSS integrated with the dead-reckoning (DR) system is also assessed.