AI Positioning Model Relative Time Input
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In 5G New Radio mobile communications, the format of model input data for AI/ML positioning models significantly impacts positioning performance and signaling overhead, necessitating a proper format for enhanced accuracy.
Innovation Solution
The proposed solution involves using relative time as model input for AI/ML positioning models, where a user equipment (UE) measures a channel delay profile with multiple path timings, adjusts these timings based on a timing difference associated with a reference network node, and then inputs these adjusted timings into the positioning model to generate accurate position estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If absolute time values are used as model input for AI/ML positioning, then the positioning model can process timing information, but the model accuracy is compromised due to timing misalignment between different network nodes
Solution Approach 1:
The patent transforms the time parameter from absolute values to relative values by applying timing differences. Specifically, the UE adjusts the first path timing using a timing difference associated with a reference network node, converting absolute timing measurements into relative timing measurements that are consistent across different network nodes. This parameter transformation resolves the timing alignment issue and enables accurate positioning.
2Measurement precision
If detailed timing adjustment information is reported to the network, then positioning accuracy improves, but signaling overhead increases
Solution Approach 1:
The patent extracts only the essential timing adjustment information (timing difference relative to reference node) that is necessary for positioning accuracy, rather than reporting complete absolute timing data for all network nodes. This extraction approach maintains positioning performance while minimizing the amount of data that needs to be transmitted to the network.
3Reliability
If multiple network nodes are used for positioning measurements, then positioning robustness improves, but timing synchronization complexity increases
Solution Approach 1:
The patent segments the timing synchronization problem by introducing a reference network node concept. Instead of requiring all network nodes to be synchronized to a common absolute time reference, the system divides the problem into node-specific relative timing adjustments. Each node's timing is adjusted relative to the reference node, simplifying the overall synchronization complexity while maintaining robustness through multiple measurement points.
Data Source
AI summary
Various solutions for an artificial intelligence/machine learning (AI/ML) positioning model using relative time input with respect to an apparatus in mobile communications are described. The apparatus may measure a first channel delay profile with a first path timing according to a first reference signal associated with a first network node. The apparatus may adjust the first path timing by a timing difference associated with a reference network node. The apparatus may: (1) generate a model output by a positioning model based on the first channel delay profile with the adjusted first path timing used as model inputs; or (2) report the first channel delay profile with the adjusted first path timing to a network.