On the Problem of Double-Filtering in PPP-RTK

Area of Science:

• Geomatics Engineering
• Satellite Navigation Systems
• Signal Processing

Background:

• Precise Point Positioning-Real Time Kinematic (PPP-RTK) systems rely on external corrections.
• Standard user-filters violate Kalman filter assumptions due to time-correlated corrections, leading to imprecise Global Navigation Satellite System (GNSS) solutions.
• Correction latency exacerbates precision loss in traditional PPP-RTK user-filters.

Purpose of the Study:

• To develop a novel multi-epoch formulation for the PPP-RTK user-filter.
• To incorporate both the uncertainty and temporal correlation of corrections into the user-filter.
• To enhance the precision of GNSS parameter solutions in PPP-RTK.

Main Methods:

• Augmenting the user-filter state-vector to include correction characteristics.
• Jointly updating user parameters and corrections within the measurement update step.
• Implementing a multi-epoch Kalman filter approach.

Main Results:

• The proposed formulation successfully incorporates correction uncertainty and temporal correlation.
• Numerical results demonstrate superior performance compared to conventional methods.
• The new filter achieves the minimum-variance property, enhancing solution precision.

Conclusions:

• The novel multi-epoch PPP-RTK user-filter formulation significantly improves GNSS solution precision.
• Accounting for correction temporal correlation and uncertainty is crucial for optimal filter performance.
• This approach offers a more robust and accurate method for real-time GNSS positioning.