RAN ML Positioning Capability Signaling for Accurate UE Location
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems face challenges in accurately determining the positioning of user equipment (UE) due to the complexity of high-frequency bands and dense deployments in 5G networks, necessitating improved methods for precise location estimation.
Innovation Solution
Implementing machine learning (ML) positioning functionalities within radio access network (RAN) nodes and network entities, where RAN nodes transmit ML positioning capabilities and SRS measurements to network entities, enabling a common understanding of ML model deployment scenarios and enhancing positioning accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning positioning functionalities are implemented in RAN nodes, then positioning accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the positioning functionality by identifying and listing specific machine learning positioning functionalities (e.g., SRS-based positioning, channel state information-based positioning) that can be independently deployed in RAN nodes. This allows the network to selectively implement only the necessary positioning functions based on deployment scenarios, reducing unnecessary complexity while maintaining accuracy where needed.
Solution Approach 2:
The patent introduces a capability indication mechanism that allows RAN nodes to dynamically indicate their machine learning positioning capabilities to network entities. This enables the system to adapt the positioning approach based on the specific capabilities available at each RAN node, allowing for optimized positioning accuracy without requiring all nodes to implement the full complex positioning suite.
2Adaptability or versatility
If multiple machine learning positioning functionalities are supported, then positioning versatility is improved, but ease of operation deteriorates
Solution Approach 1:
The patent creates a universal capability indication framework where RAN nodes can announce support for multiple machine learning positioning functionalities through a standardized mechanism. This allows the network to universally handle different positioning scenarios (SRS-based, channel state information-based, etc.) through a common interface, simplifying operations despite supporting multiple specialized functions.
Solution Approach 2:
The patent implements a feedback mechanism where RAN nodes provide capability information to network entities, which then select appropriate positioning functionalities based on this feedback. This automated selection process reduces operational complexity by eliminating manual configuration requirements, allowing the system to automatically adapt to different deployment scenarios based on node capabilities.
Data Source
AI summary
Disclosed are techniques for communication. In an aspect, a radio access network (RAN) node transmits, to a network entity, a set of machine learning positioning capabilities supported by the RAN node, wherein the set of machine learning positioning capabilities includes a list of identifiers of a set of machine learning positioning functionalities supported by the RAN node, wherein the set of machine learning positioning functionalities is associated with a set of machine learning models that support the set of machine learning positioning functionalities, and wherein each machine learning positioning functionality of the set of machine learning positioning functionalities is associated with one or more machine learning models of the set of machine learning models, and transmits, to the network entity, one or more measurements of one or more sounding reference signal (SRS) resources transmitted by a user equipment (UE).