Wireless AI/ML Capability Signaling for Low-Overhead Model Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for an efficient method to manage artificial intelligence (AI)/machine learning (ML) models and functionalities in wireless communication systems, particularly in managing signaling between terminal devices and base station devices, to optimize network performance and reduce unnecessary signaling and resource waste.
Innovation Solution
A method and apparatus are provided to efficiently manage AI/ML models and functionalities by configuring capability information and controlling signaling between terminal and base station devices, including the transmission of capability and preference information, and implementing a prohibit timer to manage AI functionality updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If AI/ML models and functionalities are frequently updated in wireless communication systems, then system performance and adaptability are improved, but signaling overhead and resource consumption increase
Solution Approach 1:
The system performs preliminary actions by having the terminal device report its AI/ML capability information (supported models, functionalities, and parameters) to the base station in advance. The base station then stores this information and uses it to determine appropriate AI/ML configurations before actual communication occurs, avoiding the need for frequent capability negotiation and reducing signaling overhead during model updates.
Solution Approach 2:
The base station creates and stores copies of the terminal device's capability information in its memory. This copied information is then referenced during AI/ML model management and configuration processes, eliminating the need for repeated capability exchange signaling and enabling efficient model updates without redundant communication.
2Manufacturing precision
If comprehensive capability information is transmitted between terminal and base station, then AI/ML model management accuracy is improved, but message size and processing complexity increase
Solution Approach 1:
The capability information is segmented into distinct components: supported AI/ML model types, supported functionalities, and model parameters. Each component is separately identified and structured in the capability message, allowing the base station to process and store specific aspects of capability information independently, reducing overall processing complexity while maintaining comprehensive accuracy.
Solution Approach 2:
The system transforms capability information into standardized parameters that can be efficiently stored and processed. The base station converts received capability data into structured parameter formats (such as supported model identifiers, functionality flags, and parameter ranges) that simplify subsequent model management operations and reduce the complexity of configuration tasks.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A terminal device may include a memory configured to store at least one artificial intelligence (AI) model or at least one AI functionality for performing wireless communication based on AI, a communication interface, and a processor configured to: configure capability information of the terminal device, the capability information including information on a number of at least one AI model or a number of at least one AI functionality, which is supportable by the terminal device, and control the communication interface to transmit a first message including the capability information of the terminal device to a base station device.