5G AI/ML Capability Signaling for UE and Network Coordination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for techniques to indicate Artificial Intelligence (AI) and/or Machine Learning (ML) capability in 3rd Generation Partnership Project (3GPP) 5G networks and User Equipment (UE), as existing systems lack effective methods for reporting and managing AI/ML capabilities between network entities and devices.
Innovation Solution
Methods and apparatus are provided for indicating UE AI/ML capability to a 3GPP 5G network and vice versa, using signaling protocols such as RRC and NG, with specific information elements (IEs) for transmitting UE and network AI/ML capability indications, including model IDs and operation details.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing signaling protocols are used without AI/ML capability indication, then system compatibility is maintained, but AI/ML functionality cannot be effectively managed or utilized
Solution Approach 1:
The patent embeds AI/ML capability indication information elements within existing 3GPP signaling protocols (RRC, NG, etc.). The capability indication is nested as additional information fields within standard signaling messages, allowing AI/ML functionality to be indicated without creating separate protocol structures. This maintains system compatibility while enabling new capabilities.
Solution Approach 2:
The patent creates universal information elements that can indicate various AI/ML capabilities across different network entities and devices. These IEs are designed to be multi-functional, supporting indication of model IDs, operation types, and capability levels in a unified manner that works across RRC, NG, and other signaling interfaces.
2Measurement precision
If detailed AI/ML capability information is transmitted, then capability management precision is improved, but signaling overhead increases
Solution Approach 1:
The patent segments AI/ML capability information into distinct information elements including model ID, operation type, and capability level. Each IE carries a specific aspect of capability information, allowing precise indication while organizing data efficiently. The segmentation enables selective transmission of only relevant capability details rather than transmitting all possible information.
Solution Approach 2:
The patent uses parameter-based indication where capability details are conveyed through structured parameters within information elements. By changing the parameter structure to include specific fields for model identification, operation types, and capability levels, the system achieves precise capability management while maintaining compact data representation that controls signaling overhead.
Data Source
AI summary
The disclosure relates to a 5G or 6G communication system for supporting a higher data transmission rate. There is disclosed a first method for reporting user equipment (UE) artificial intelligence (AI)/machine learning (ML) capability to a network. The first method comprises: transmitting, to the network, an indication of the UE AI/ML capability. There is also disclosed a second method for reporting network AI/ML capability to a UE. The second method comprises: transmitting, to the UE, an indication of the network AI/ML capability.