AI Node Channel Feature Extraction for UE Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AI models for determining the location of user equipment (UE) in wireless communication systems face challenges due to high complexity and accuracy issues, particularly in non-line-of-sight environments, and require efficient methods to process large amounts of channel data while reducing signaling overhead.
Innovation Solution
A method involving the extraction of feature information from channel data between UE and base stations using signature transforms, data augmentation, and compression, which includes weighting time and energy of multipath signals, and utilizing AI models to determine UE location and monitor model updates based on channel characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing AI models are used for UE positioning, then positioning functionality is provided, but the model complexity is high and positioning accuracy is insufficient
Solution Approach 1:
The patent extracts and transmits only the essential channel information (channel impulse response or channel frequency response) from the complex channel data to the AI model, rather than transmitting all raw channel data. This extraction approach reduces the input data dimensionality while preserving the most relevant positioning features, thereby improving positioning accuracy without requiring excessively complex models.
Solution Approach 2:
The patent segments the channel information into distinct features (such as multipath components, time of arrival, angle of arrival) and processes them separately through the AI model. This segmentation allows the model to focus on specific positioning-relevant characteristics rather than processing the entire channel data matrix, reducing complexity while maintaining or improving positioning precision.
2Measurement precision
If existing AI models process channel data, then positioning is performed, but the amount of input data is large causing large feedback overhead
Solution Approach 1:
The patent extracts only the necessary channel features (such as dominant multipath components, time delays, and energy values) required for accurate positioning, rather than transmitting the complete channel data set. This selective extraction significantly reduces the feedback overhead while preserving the information most critical for positioning accuracy.
Solution Approach 2:
The patent transforms the channel information from its original high-dimensional form into a reduced set of key parameters (such as time of arrival, angle of arrival, and signal strength of dominant paths). This parameter transformation maintains the essential positioning information while dramatically reducing the data volume that needs to be processed and transmitted.
Data Source
AI summary
A node in a wireless communication system and a method performed by the same are provided. The method includes obtaining information related to a channel between a user equipment (UE) and a base station, and extracting first feature information based on the information related to the channel between the UE and the base station, the first feature information being used to determine information of a location of the UE using an artificial intelligence (AI) model and/or to monitor whether the AI model needs to be updated.