Wireless Device Positioning with High-Elevation SV Filtering

The patent applies local quality by differentiating between SVs with line-of-sight signals and those with non-line-of-sight signals. It selectively processes measurements from SVs based on their individual signal qualities, using measurements from SVs with clear line-of-sight paths while excluding or down-weighting measurements from SVs blocked by obstructions. This localized quality assessment per SV resolves the contradiction by maintaining accuracy through selective measurement usage while preserving coverage through systematic evaluation of all available SVs.

The patent changes the parameter of measurement reliability by introducing quality indicators and confidence levels for each SV measurement. It dynamically adjusts the weighting and selection of measurements based on detected signal obstruction parameters, such as comparing current measurements with historical data to identify NLOS conditions. This parameter transformation allows the system to maintain positioning coverage while improving accuracy by relying on high-quality measurements.

The patent transforms the selection criterion from a single parameter (elevation angle) to a composite parameter that includes elevation, azimuth, signal strength, and historical obstruction patterns. By changing the selection parameters to a multi-dimensional criterion, the system can identify SVs that provide accurate measurements not just based on high elevation, but based on overall signal quality and geometric diversity, thus maintaining sufficient number of SVs while improving measurement accuracy.

The patent applies partial action by selecting only the necessary subset of SVs that provide high-quality measurements rather than requiring a fixed minimum number of SVs. It uses quality thresholds and confidence levels to determine the optimal number of SVs to use, which may vary dynamically based on environmental conditions. This allows the system to use fewer SVs when quality is high, improving accuracy, while still maintaining adequate coverage when more SVs are available.

The patent applies preliminary action by pre-processing and pre-classifying SV signals into LOS and NLOS categories using historical data and initial quality assessments. It establishes baseline characteristics of reliable SVs before the actual positioning calculation, so that during real-time operation, the system can quickly reference pre-computed quality indicators rather than performing full signal analysis. This preliminary classification reduces computational complexity while maintaining accuracy improvement from NLOS avoidance.

The patent implements feedback mechanisms where positioning results and signal quality measurements are continuously monitored and used to refine the identification of NLOS conditions. The system uses feedback from previous positioning attempts to adjust its NLOS detection thresholds and SV selection criteria, making the identification process more efficient over time. This feedback loop improves accuracy while reducing complexity by learning from past experiences rather than requiring complex real-time analysis for every measurement.