Terminal AI Model Control for 5G Mobility and Power Saving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI models in 5G networks consume excessive power and may produce unreasonable or imprecise outputs, degrading user experience due to poor generalization performance and inefficiencies in network management.
Innovation Solution
Implement methods for deactivating or reducing AI model capabilities in terminal devices based on specific events such as signal measurement criteria, overheating, radio link failures, or state changes, and transmitting measurement reports to network devices for improved mobility management and power saving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If AI models are continuously activated for network management, then network optimization capabilities are improved, but power consumption increases
Solution Approach 1:
The patent implements dynamic activation and deactivation of AI models based on terminal device states. The AI model is activated when the terminal is in connected state and deactivated when entering idle or inactive state, allowing the system to adapt its computational resources to actual operational needs rather than maintaining continuous operation.
Solution Approach 2:
The system employs periodic state transitions (connected -> idle/inactive) to control AI model operation. By monitoring terminal device states and periodically activating/deactivating the AI model accordingly, the system achieves energy-efficient operation while maintaining optimization capabilities when needed.
2Measurement precision
If AI models operate with high computational capabilities, then network management precision is improved, but device overheating increases
Solution Approach 1:
The patent dynamically adjusts AI model operation based on terminal device thermal conditions and state transitions. When the terminal enters idle or inactive state, the AI model is deactivated to reduce computational heat generation, while maintaining precision capabilities during connected state when thermal conditions permit.
Solution Approach 2:
The system converts the potential harm of thermal accumulation into a beneficial control mechanism by using state transitions (particularly entering idle/inactive state) as triggers to deactivate the AI model, thereby preventing overheating while still enabling high-precision operations when thermally appropriate.
3Productivity
If AI models are activated during handover, then mobility management is improved, but system complexity increases
Solution Approach 1:
The patent prepares for handover scenarios by pre-configuring AI model behavior. When handover is detected or anticipated, the system deactivates the AI model in advance or adjusts its operation to avoid complex computations during the transition period, simplifying the handover process while maintaining mobility management effectiveness.
Solution Approach 2:
The system dynamically adapts AI model operation during handover events by deactivating the model when entering idle or inactive state during the transition, thereby reducing computational complexity during this critical period while still enabling efficient mobility management through state-based control.
Data Source
AI summary
Embodiments of the present disclosure relate to methods, devices and computer readable media of communication. In one aspect, in accordance with a determination that an event occurs, a terminal device deactivates or reduces capabilities of an AI model. The event comprising at least one of the following: a signal measurement criterion is met; an overheating condition is detected; a RLF is detected; a first indication indicating the deactivating or reducing is received; the terminal device enters an idle or inactive state; or a handover to a target cell is to be performed, an AI model associated with the target cell being specific to a cell. In this way, management of AI model may be achieved and power saving may be attained.