AI/ML Capability Reporting in Mobile Communication Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile communication systems lack a clear method for identifying and managing artificial intelligence (AI) and machine learning (ML) functionalities/models between base stations and user equipment (UEs), leading to inefficiencies in signaling and resource allocation.
Innovation Solution
A method and apparatus for identifying AI/ML functionalities/models through an AI/ML-related capability reporting procedure and model information reporting, where dataset identification information is shared between base stations and UEs, and life cycle management (LCM) techniques are selectively applied based on the number of AI/ML models supported.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If AI/ML functionalities are implemented in mobile communication systems without clear identification methods, then the system can support advanced processing capabilities, but signaling efficiency deteriorates and resource allocation becomes inefficient
Solution Approach 1:
The capability reporting is segmented into multiple stages: initial capability indication, detailed capability reporting, and model information reporting. This segmentation allows the system to support AI/ML functionalities while controlling signaling overhead by only exchanging detailed information when necessary.
Solution Approach 2:
The base station performs preliminary capability inquiry before actual AI/ML operations. The UE provides initial capability indication in advance, allowing the network to prepare appropriate configuration and avoid unnecessary detailed signaling exchanges.
2Reliability
If comprehensive AI/ML capability information is reported between base station and UE, then the system can effectively identify and manage supported functionalities, but the complexity of the identification procedure increases
Solution Approach 1:
The capability reporting procedure is made dynamic and adaptive. The base station selectively requests different levels of detail based on initial capability indications. The UE adapts its reporting content based on network requirements and its own capability status, reducing unnecessary procedural complexity while maintaining identification accuracy.
Solution Approach 2:
Instead of requiring complete capability information in all cases, the system uses partial action by implementing staged reporting. The UE provides essential capability information initially, and only reports detailed model information when specifically requested and when it adds value to the communication process.
3Adaptability or versatility
If multiple AI/ML models are managed between base station and UE, then the system can support diverse machine learning applications, but resource allocation and lifecycle management become more difficult
Solution Approach 1:
Model lifecycle management is segmented into distinct phases: model capability indication, model information reporting, model activation, and model updates. Each phase is handled through specific signaling procedures, making the management of multiple diverse models more systematic and easier to operate.
Solution Approach 2:
The system implements feedback mechanisms where the UE reports supported model types and capabilities to the base station, which then provides appropriate configuration and activation commands. This feedback loop enables efficient resource allocation and lifecycle management of multiple AI/ML models based on actual network needs and UE capabilities.
Data Source
AI summary
A method of identifying an artificial intelligence (AI)/machine learning (ML) functionality and model supported for mobile communication operated in a mobile communication system including a base station and one or more user equipments (UEs), the method comprising: delivering, from the base station, dataset identification information regarding at least one dataset to the UE; and reporting, by the UE, valid AI/ML-related UE capability for the at least one dataset corresponding to the dataset identification information to the base station.