Positioning Environment Modeling With Simulated Measurement Training

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing wireless communication systems, particularly 5G NR, face challenges in improving positioning accuracy and efficiency due to limited access to diverse environments for data collection, leading to suboptimal training of positioning models.

Innovation Solution

A method and apparatus that simulate positioning measurements based on environment information, combining simulated and measured data to train AI/ML positioning models, enabling training in inaccessible or difficult-to-reach areas.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If positioning models are trained using only measured data from accessible areas, then data collection is simple and cost-effective, but positioning accuracy in diverse environments deteriorates due to limited environmental coverage

Engineering Contradiction:
Improvepositioning accuracyVSAvoidenvironmental coverage
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent creates virtual copies of physical environments through 3D modeling and digital twins. These virtual environment models replicate real-world positioning scenarios, allowing the system to generate synthetic training data that mirrors actual measurement conditions without requiring physical access to diverse locations. This copying approach enables comprehensive environmental coverage while maintaining data authenticity.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The system performs preliminary actions by pre-collecting environment information and pre-generating synthetic positioning measurements for various scenarios before actual positioning needs arise. This advance preparation creates a library of training data covering diverse environments, so when positioning is needed in any environment, the model is already trained and ready, eliminating the need for on-site data collection.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If measured positioning data is collected from diverse environments, then positioning model training improves, but access to inaccessible areas becomes a limiting factor

Engineering Contradiction:
Improvemodel training qualityVSAvoiddata collection accessibility
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

Instead of physically accessing inaccessible areas to collect measurements, the system creates virtual copies of these environments using available information such as building plans, maps, or partial measurements. The synthetic data generation process replicates positioning measurements as if they were taken in the actual inaccessible environment, bypassing the need for physical access while maintaining training quality.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent introduces an intermediary synthetic data generation process that bridges the gap between accessible and inaccessible areas. This intermediary layer uses environment information and simulation algorithms to translate available data into comprehensive training datasets, mediating between what can be easily measured and what is needed for robust model training.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If synthetic positioning measurements are generated based on environment information, then data availability improves for training, but measurement precision may deteriorate compared to real measurements

Engineering Contradiction:
Improvedata generation efficiencyVSAvoidmeasurement accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary calibration by generating synthetic measurements using simulated positioning signals that are pre-configured to match real signal characteristics. Environment information such as building geometry, material properties, and antenna configurations are incorporated into the simulation to ensure synthetic measurements accurately reflect real-world physics and measurement conditions before actual positioning operations begin.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback mechanisms where synthetic measurements are continuously refined based on comparison with available real measurements. The system uses measured data to validate and adjust simulation parameters, creating a feedback loop that progressively improves measurement precision. This feedback ensures synthetic data maintains high accuracy while preserving the efficiency benefits of generation.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS12607705B2Positioning environment simulation based upon modeling
Publication Date: 2026.04.21 QUALCOMM INC
  • US12607705B2 patent drawing
  • US12607705B2 patent drawing
  • US12607705B2 patent drawing

AI summary

A wireless device may receive environment information associated with an area. The wireless device may simulate a set of positioning measurements based on the environment information. The wireless device may calculate a positioning environment based on the simulated set of positioning measurements. The wireless device may calculate the positioning environment further based on a set of measured positioning signals obtained by the wireless device. The wireless device may output the positioning environment to train a positioning model. The wireless device may output the positioning environment by training the positioning model at the wireless device based on the positioning environment. The wireless device may output the positioning environment by transmitting the positioning environment to a training entity to train the positioning model.