AI/ML Model Activation in 5G UE and Network Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 5G mobile communication systems face challenges in managing and training Artificial Intelligence (AI) and Machine Learning (ML) models efficiently, particularly in handling model distribution, activation, and management across diverse user equipment (UE) types and network environments.
Innovation Solution
A method for UE to transmit information on supported AI/ML models, receive activation instructions, and activate these models based on received information, with network entities managing model distribution and activation through messages like INITIAL CONTEXT SETUP and UE CONTEXT MODIFICATION, utilizing entities such as AMF and NG-RAN to handle model lists and profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple AI/ML models are supported across diverse UE types and network environments, then model functionality and versatility are improved, but system complexity increases
Solution Approach 1:
The patent segments the AI/ML model management system into distinct functional modules: model information transmission (UE), model distribution management (AMF), model activation control (NG-RAN), and model execution (UE). This segmentation allows each component to handle specific tasks independently, reducing overall system complexity while supporting multiple model types across diverse UE configurations
Solution Approach 2:
The patent implements a universal model management framework where the AMF handles distribution for all UE types, the NG-RAN manages activation across different network environments, and the UE can execute multiple model types (classification, regression, clustering, etc.). This multi-functional design enables the system to support diverse AI/ML applications without requiring separate management mechanisms for each model type
2Productivity
If AI/ML models are distributed and activated across the network, then model usage efficiency is improved, but management overhead increases
Solution Approach 1:
The patent implements preliminary actions by having the UE transmit model information (including model identifiers, types, and parameters) to the AMF in advance before actual model execution. The AMF pre-configures model distribution policies and the NG-RAN pre-arrives activation conditions based on network environment assessments. This preliminary configuration reduces real-time management overhead during actual model deployment and execution
Solution Approach 2:
The patent introduces the AMF as an intermediary between the UE and NG-RAN for model distribution management, and the NG-RAN as an intermediary between the core network and UE for model activation control. These intermediary entities handle the complexity of model management tasks, shielding the UE from management overhead while improving model usage efficiency through centralized coordination
Data Source
AI summary
The disclosure relates to a 5G or 6G communication system for supporting a higher data transmission rate. A UE transmits information on at least one first artificial intelligence (AI)/machine learning (ML) model, wherein the information on the at least one first AI/ML model includes a list of AI/ML models, and wherein an AI/ML model included in the list of AI/ML models is identified by a first AI/ML model identifier (ID), receives at least one AI/ML model information indicating at least one AI/ML model to be activated, wherein the at least one AI/ML model to be activated is identified by a second AI/ML model ID; and activates the indicated at least one AI/ML model based on the received at least one AI/ML model information.