Differential Kalman Filter Design for GNSS Open Loop Tracking

Abstract

Global navigation satellite system (GNSS) positioning in an urban environment is in need for accurate, reliable and robust positioning. Unfortunately, conventional closed-loop tracking fails to meet the demand. The open loop tracking shows improved robustness, however, the precision is unsatisfactory. We propose a differential Kalman filter for open loop, of which the measurement vector contains the differential values of open loop navigation results between adjacent epochs. The differential Kalman filter makes use of the satellite geometry (i.e., spatial domain) and motion relationship (i.e., temporal domain) to filter frequency and code phase estimations of conventional open loop tracking. The improved performances of this architecture have been analyzed theoretically and demonstrated by road tests in an urban environment. The proposed architecture shows more than 50% accuracy improvement than the conventional open-loop tracking architecture.