Optimize Kalman Filter For Smart Phone Location Tracking Using GNSS Measurements

Abstract

Point precise positioning (PPP) method allows to locate user with Global Navigation Satellite System under good condition. But this method is quite challenging to locate user when user is in urban areas with obstacles. Another alternative solutions such as Real Time Kinematic solve these problems with differential measurements. In real world scenario, raw GNSS measurements have usually noisy and uncertain data. Kalman filter is successful method to optimally update variables with indirect measurements. This research proposes the method to optimize Kalman filter based PPP to improve smart phone location tracking. Evaluation method showed that proposed method worked well with approximately 3.0 meter accuracy and outperformed base line model which is calculated using Weighted-Least-Square (WLS) solver.