Positioning Signal Measurement Filtering for Wireless Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication systems, particularly 5G NR, face challenges in improving positioning accuracy and reliability due to limitations in data quality and conditioning for training and verifying positioning models.
Innovation Solution
The proposed solution involves a method where user equipment (UE) and network entities configure conditional configurations to select and filter subsets of positioning signal measurements for training or verifying positioning models. This ensures high-quality data collection by adhering to specific conditions related to measurement types, labeling methods, and quality standards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If positioning models are trained using all available measurements, then the quantity of training data increases, but the quality and reliability of the positioning model deteriorates due to inclusion of low-quality or erroneous measurements
Solution Approach 1:
The patent extracts and filters out low-quality or erroneous measurements from the complete set of positioning measurements. By identifying and removing problematic data points that would degrade model performance, the system retains only high-quality measurements for training, thus resolving the contradiction between having sufficient data quantity and maintaining data quality.
Solution Approach 2:
The patent applies different quality thresholds and filtering criteria to different types of measurements and positioning scenarios. Instead of uniform treatment, the system evaluates each measurement's quality individually using multiple criteria (signal strength, geometric accuracy, temporal validity) and selectively includes only those meeting local quality standards, thereby improving overall model reliability while maintaining adequate training data volume.
2Productivity
If positioning models are trained using all available measurements, then the productivity of data collection increases, but the measurement precision and positioning accuracy deteriorate due to inclusion of erroneous data
Solution Approach 1:
The patent performs preliminary filtering and quality assessment actions on measurements before they are used for training. By pre-screening measurements against defined criteria (signal quality thresholds, geometric dilution of precision limits, temporal consistency checks), the system ensures that only precision-worthy data enters the training pipeline, maintaining both high productivity and high measurement precision.
Solution Approach 2:
The system incorporates feedback mechanisms where positioning model performance metrics are continuously monitored and fed back into the data collection and filtering process. This feedback loop allows the system to adjust quality thresholds and filtering parameters dynamically, ensuring that data collection productivity remains high while measurement precision and positioning accuracy are maintained through iterative optimization.
3Reliability
If conditional configurations are implemented for filtering measurements, then the reliability of positioning models improves, but the device complexity increases due to additional filtering and verification processes
Solution Approach 1:
The patent implements a universal measurement filtering framework that handles multiple quality assessment functions through a single integrated system. The same filtering mechanism processes various measurement types (signal strength, timing, geometric parameters) using unified quality criteria, reducing the need for separate complex filtering systems for each measurement type and thereby limiting the increase in device complexity while maintaining high reliability.
Solution Approach 2:
The system manages device complexity by dynamically adjusting filtering parameters and thresholds based on operational conditions rather than using fixed complex rules. Quality criteria such as signal strength thresholds, geometric accuracy limits, and temporal validity windows are parameterized and can be modified based on current positioning scenarios, allowing the system to maintain high reliability through adaptive filtering while keeping the underlying complexity manageable through parameterization rather than structural complexity.
Data Source
AI summary
A user equipment (UE) may receive a configuration for collecting measurements to at least one of train or verify a positioning model. The UE may receive a set of positioning signals. The UE may measure the set of positioning signals. The UE may output a subset of the measured set of positioning signals to at least one of train or verify the positioning model based on the configuration. The UE may output the subset of the measured set of positioning signals by training verifying the positioning model at the UE based on the subset of the measured set of positioning signals, or by transmitting the subset of the measured set of positioning signals to at least one of train or verify the positioning model.