Cloud-Assisted Satellite Positioning with Prediction-Noise Weighting

The patent converts the harmful noise in satellite signals into useful information by calculating prediction noise values. These noise predictions are then used to determine weights for each satellite signal, where signals with lower predicted noise receive higher weights. This transforms the previously harmful noise characteristic into a beneficial weighting mechanism that improves overall positioning accuracy through weighted least squares calculation.

The patent introduces a new parameter dimension for signal evaluation by calculating prediction noise values for each satellite signal. This parameter change transforms the traditional equal-weight treatment of all satellite signals into a differentiated weighting system where each signal's reliability is quantified through its predicted noise level. The weighting parameter is dynamically adjusted based on signal quality metrics, enabling more reliable position calculations.

The patent applies local quality by assigning different weights to different satellite signals based on their individual prediction noise values. Instead of treating all satellite signals uniformly, each signal receives a localized weight that reflects its specific quality characteristics. This allows the system to maintain computational efficiency while achieving higher accuracy by giving more importance to high-quality signals and less to low-quality ones.