Positioning performances of single frequency GPS, GLONASS and carrier based algorithms in a software platform /

Savaş, Caner
There exist various algorithms (signal acquisition, tracking, etc.), methods (RTK (Real Time Kinematic), DGPS (Differential Global Positioning System), etc.) or aid systems (SBAS (Satellite Based Augmentation Systems), INS (Inertial Navigation System) aid etc.) to improve the GNSS (Global Navigation Satellite System) positioning accuracy. The positioning accuracy of different systems is usually determined by using a test setup which includes costly hardware and software units. In order to compare positioning performances of GPS, GLONASS and carrier based algorithms, namely LAMBDA (Least-squares Ambiguity De-correlation Adjustment) method, in a cost effective manner, a software platform is developed. GNSS SDR (Software Defined Radio) receiver approach provides ease in modification of algorithms to meet the expectations of different types of applications. In the light of this approach, single L1 frequency GPS and GLONASS SDR algorithms are implemented in MATLAB. IF (Intermediate Frequency) sampled data is taken from the commercial IFEN GNSS receiver to test the algorithms. In order to improve the positioning accuracy, GPS-GLONASS combined solution and two receiver positioning algorithms, namely LAMBDA method, for GPS L1 single frequency approach are implemented. It is observed that GLONASS solution is in general less accurate than GPS solution. Furthermore, GPS-GLONASS combined solution approaches to GPS solution which reveals the fact that combined solution is useful when some GPS satellites are not observable. Moreover, a tracking lock status indicator in signal tracking loops is implemented to improve the accuracy for dynamic scenario. Finally, it is shown that LAMBDA method can achieve centimeter level accuracy in two-receiver configuration.