Low-Complexity Adaptive Direct-State Kalman Filter for Robust GNSS Carrier Tracking

Abstract

This paper evaluates the implementation of a low-complexity adaptive direct-state Kalman filter (DSKF) for robust carrier phase tracking of global navigation satellite system (GNSS) signals. This architecture consists of a loop-bandwidth control algorithm (LBCA)-based lookup table (LUT)-DSKF in an FLL-assisted-PLL (FAP) tracking scheme. The FAP considers the carrier phase and frequency Doppler measurements to achieve a robust tracking. The use of the DSKF in the FAP achieves optimal performance, assuming a known Gaussian distributed model of the states and the measurements. However, the performance decays in time-varying scenarios where the measurements' distribution changes due to noise, signal dynamics, multi-path, and non-line-of-sight effects. In addition, the DSKF's implementation in real-time applications requires a high computational cost. This study derives the so-called LUT-DSKF for the FAP, a simplified DSKF that considers the convergence of the Kalman gains. In addition, the LBCA is used to adapt the time of response of this architecture and improve the tracking performance in time-varying scenarios. The presented technique is compared with the adaptive LUT-DSKF in a phase-locked loop (PLL) tracking scheme. These two tracking architectures are implemented in an open software interface GNSS hardware receiver. The receiver is evaluated in simulated scenarios with different dynamics and noise cases for each implementation. The results confirm that the LBCA-based LUT-DSKF in the FAP exhibits superior dynamic tracking performance than the adaptive PLL while maintaining similar static tracking performance and low complexity.