AI Positioning Input Selection for Reduced Base Station Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The large input dimension of AI models for positioning in wireless communication networks, particularly in NR systems, leads to processing and storage burdens, and additional signaling demands, especially when transmitting positioning measurement results from base stations to terminal devices.
Innovation Solution
The method involves selecting a preset number X of base stations, where X is less than the total number M of participating base stations, to determine the positioning measurement results as the input for the AI model, reducing the input dimension and signaling burden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If positioning measurement results from all base stations are used as input to the AI model, then positioning precision is improved, but processing burden and storage requirements increase significantly
Solution Approach 1:
The patent extracts only the necessary positioning measurement results from all base stations and selects a subset to use as AI model input. By taking out only the essential data rather than processing all available data, the system maintains positioning precision while significantly reducing processing burden and storage requirements.
Solution Approach 2:
The patent segments the complete set of positioning measurement results into a manageable subset. Instead of processing all base station measurements as a single large dataset, the system divides and selects only the necessary portions, making the input dimension more manageable while preserving positioning accuracy.
2Measurement precision
If positioning measurement results from all base stations are transmitted to the terminal device, then positioning accuracy is improved, but signaling overhead increases
Solution Approach 1:
The patent extracts only the necessary positioning measurement results from all base stations and transmits only this selected subset to the terminal device. This extraction approach maintains positioning accuracy by preserving essential measurement data while significantly reducing the quantity of data transmitted, thereby lowering signaling overhead.
Solution Approach 2:
The patent applies partial action by transmitting only a subset of positioning measurement results rather than all available data. This partial transmission is sufficient to achieve accurate positioning without the excessive signaling overhead that would result from transmitting complete measurement data from all base stations.
3Measurement precision
If a large number of base stations are used for positioning measurement, then positioning precision is improved, but the input dimension of the AI model becomes very large
Solution Approach 1:
The patent extracts only the necessary positioning measurement results from all base stations and selects a subset to use as AI model input. By taking out only the essential data rather than processing all available data, the system maintains positioning precision while significantly reducing processing burden and storage requirements.
Solution Approach 2:
The patent changes the parameter of input dimension size by selecting a subset of base stations rather than using all available base stations. This parameter change reduces the input dimension to a manageable level while maintaining positioning precision through careful selection of the subset.
Data Source
AI summary
A method for determining an input of an artificial intelligent (AI) model is provided. An output of the AI model is configured to determine a position of a terminal device, and the method is performed by the terminal device and includes: obtaining positioning measurement results corresponding to a preset number X of base stations, where the preset number X is less than a total number M of all base stations participating in a positioning measurement, where X and M are both positive integers; and determining the positioning measurement results corresponding to the X base stations as the input of the AI model.