RAN UE Positioning Architecture for AI/ML Location Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication systems, particularly in NG-RAN, lack a defined procedure to access user equipment (UE) location information, which is crucial for artificial intelligence and machine learning operations, and existing methods like A-GNSS may not work reliably indoors or provide consistent data, impacting AI/ML model quality and being limited by user consent and GPS dependency.
Innovation Solution
The proposed solution enhances the UE positioning architecture by allowing the RAN to collect UE location data when needed, using intelligence about UE positioning capabilities to select appropriate location procedures, such as RAN Triggered MO-LR UE based, MTLR UE assisted, or MT-LR NR-RAN assisted, and introduces new signaling protocols like NRPPa to request and obtain UE location information, enabling the RAN to access detailed coordinates for AI/ML operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If A-GNSS positioning method is used, then positioning can be provided outdoors, but it does not work reliably indoors and is limited by user consent and GPS dependency
Solution Approach 1:
The patent implements multiple positioning methods (A-GNSS, OTDOA, ECID, AoA, AoD) within a single positioning system, allowing the system to adapt to different environments (indoor/outdoor) and conditions (user consent, GPS availability). The RAN apparatus can select appropriate positioning methods based on UE capability information and deployment conditions, making the system universally applicable across various scenarios.
Solution Approach 2:
The system dynamically changes positioning parameters by selecting different positioning methods and modes (UE-based, UE-assisted, RAN-assisted) based on UE capability information, deployment conditions (indoor/outdoor), and service requirements. This parameter adaptation resolves the contradiction by adjusting the positioning approach to match environmental conditions.
2Loss of information
If the RAN triggers location procedures based on AI/ML functions, then accurate location data is obtained for AI/ML operations, but new signaling protocols and procedures must be defined
Solution Approach 1:
The patent performs preliminary actions by defining and establishing new signaling protocols (NRPPa, LCSAP, RRC) and location procedures before the RAN can effectively trigger positioning for AI/ML operations. The framework pre-configures the system with capability information exchange mechanisms, QoS parameter definitions, and procedure selection logic, enabling subsequent accurate location data acquisition without ad-hoc protocol development.
3Adaptability or versatility
If multiple positioning methods are supported, then positioning versatility is improved, but capability information management and procedure selection complexity increases
Solution Approach 1:
The patent introduces an intermediary capability information management framework that mediates between multiple positioning methods and the RAN triggering mechanism. The system collects, manages, and standardizes UE positioning capability information, then uses this intermediary layer to automatically select appropriate positioning methods based on capabilities, deployment conditions, and QoS requirements, reducing the complexity burden on the RAN.
4Measurement precision
If location procedures are triggered on-demand for AI/ML functions, then location data accuracy is improved, but signaling overhead and processing time increase
Solution Approach 1:
The system performs preliminary actions by pre-collecting and storing UE positioning capability information, pre-configuring QoS parameters, and pre-establishing the framework for procedure selection. This preliminary preparation enables faster on-demand triggering of location procedures for AI/ML functions, reducing the actual positioning time while maintaining accuracy.
Solution Approach 2:
The patent implements dynamic procedure selection that adapts to real-time conditions. The RAN apparatus dynamically chooses positioning methods and modes based on current UE capabilities, deployment conditions, and service requirements, optimizing the balance between location accuracy and positioning time for each specific AI/ML operation context.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided an apparatus comprising means for determining, by the apparatus, that position information of a user equipment is needed for one or more functions to be performed at the apparatus; and means for triggering a location procedure for obtaining positioning information of the user equipment, in response to the determining that position information of the user equipment is needed.