Base Station Dynamic Model Configuration via Real-Time UE Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for model training and inference in 5G wireless communication networks often lead to inefficient configurations due to dynamic changes in UE capability and requirements, resulting in suboptimal performance and resource utilization.
Innovation Solution
A method where a base station requests and obtains real-time UE capability and requirement information, allowing it to dynamically determine and adjust model training and inference schemes to match the current capabilities and needs of the UE.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the base station determines model training/inference capability based on static UE hardware capability, then the system complexity is reduced, but the model processing efficiency and resource utilization deteriorate due to dynamic capability changes
Solution Approach 1:
The patent implements dynamic capability assessment by introducing real-time capability indication information that reflects the current state of UE resources (CPU, memory, battery, temperature). This allows the model processing system to adapt to dynamic changes in UE capability rather than relying on static hardware specifications, thereby improving model processing efficiency while maintaining manageable system complexity through standardized signaling mechanisms.
Solution Approach 2:
The patent establishes a feedback mechanism where the UE reports real-time capability status to the base station through uplink signaling. The base station uses this feedback information to make informed decisions about model training and inference task allocation. This closed-loop feedback system enables the network to optimize resource utilization based on actual UE conditions without requiring complex continuous monitoring infrastructure.
2Adaptability or versatility
If the base station allocates model training/inference tasks without considering real-time UE requirements, then the signaling overhead is reduced, but the adaptability to user needs deteriorates
Solution Approach 1:
The patent extracts only the essential real-time requirement parameters needed for model processing decisions (such as latency requirements, accuracy requirements, and privacy requirements) rather than transmitting complete UE state information. This selective extraction approach enables the system to achieve high adaptability to user requirements while keeping signaling overhead minimal by including only the most critical parameters in the capability indication information.
3Reliability
If the UE continuously reports detailed capability status, then the model processing adaptability improves, but the energy consumption and processing burden on UE increase
Solution Approach 1:
The patent implements partial reporting where the UE provides capability indication information at selective intervals or only when significant changes occur, rather than continuously reporting all capability parameters. The base station requests capability information only when needed for model processing decisions. This partial action approach maintains model processing reliability by providing sufficient information for accurate task allocation while significantly reducing UE energy consumption and processing burden compared to continuous detailed reporting.
Data Source
AI summary
A model processing method based on a user equipment (UE) capability, performed by a base station, includes: sending a request message to a UE. The request message is used for requesting at least one of UE hardware capability information, real-time UE capability information, or real-time UE requirement information for a model. In addition, the method includes: obtaining feedback information sent, based on the request message, by the UE. The feedback information includes information requested by the request message. Furthermore, the method includes: determining at least one of a model training scheme or a model inference scheme based on the feedback information, to train the model based on the model training scheme, or perform an inference on the model based on the model inference scheme or train the model based on the model training scheme and perform the inference on the model based on the model inference scheme.