Performance Evaluation of Adaptive Tracking Techniques with Direct-State Kalman Filter

Area of Science:

• Signal Processing
• Navigation Systems
• Estimation Theory

Background:

• Modern Global Navigation Satellite System (GNSS) receivers utilize the direct-state Kalman filter (DSKF) for optimal state estimation under Gaussian assumptions.
• Time-varying scenarios in GNSS, including noise and multipath, degrade DSKF performance by necessitating difficult-to-model noise parameters, leading to sub-optimal solutions.
• Adaptive tracking techniques are crucial for maintaining DSKF performance in dynamic and noisy environments.

Purpose of the Study:

• To evaluate the performance of robust adaptive tracking techniques for direct-state Kalman filters (DSKF) in GNSS receivers.
• To introduce and assess two novel methods, LBCA-based DSKF and LBCA-based lookup table (LUT)-DSKF, for adapting DSKF using the loop-bandwidth control algorithm (LBCA).
• To compare the proposed adaptive techniques against the carrier-to-noise density ratio (C/N0)-based DSKF in simulated GNSS scenarios.

Main Methods:

• Implementation of two LBCA-based adaptive DSKF techniques: one adapting steady-state process noise variance and the other relating loop bandwidth to Kalman gains.
• Integration of these adaptive techniques into an open software interface GNSS hardware receiver.
• Performance evaluation through simulated scenarios featuring diverse dynamics and noise conditions, assessing receiver tracking and overall system performance.

Main Results:

• The loop-bandwidth control algorithm (LBCA) effectively adapts the DSKF in time-varying GNSS scenarios.
• The LBCA-based lookup table (LUT)-DSKF demonstrates superior static and dynamic system performance compared to existing adaptive DSKF methods.
• The LBCA-based LUT-DSKF achieves the lowest implementation complexity among the evaluated adaptive tracking techniques.

Conclusions:

• The loop-bandwidth control algorithm (LBCA) is a viable and effective method for adapting direct-state Kalman filters (DSKF) in GNSS receivers.
• The LBCA-based LUT-DSKF offers a significant advancement in GNSS receiver tracking performance, balancing robustness, adaptability, and computational efficiency.
• This research validates the potential of LBCA-driven adaptation for enhancing the reliability and accuracy of satellite navigation systems.