AI Model Transmission Screening for Wireless Compatibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication systems lack efficient methods for transmitting and receiving artificial intelligence (AI) models, leading to high resource overhead and potential compatibility issues between base stations and user equipment due to mismatched AI models, which affects the performance and efficiency of AI-based communication tasks.
Innovation Solution
The proposed solution involves transmitting AI model attributes and specific AI models separately, allowing for preliminary screening of suitable models based on receiver capabilities, reducing resource overhead and selecting optimal transmission modes, and providing feedback on actual model execution status to ensure proper functioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If AI models are transmitted without preliminary screening based on receiver capabilities, then all possible models can be made available, but resource overhead increases and compatibility issues occur
Solution Approach 1:
The patent applies preliminary action by transmitting model attribute information (such as model type, complexity, and resource requirements) before transmitting the actual AI models. The receiver evaluates these attributes against its capabilities to determine which models to download, thereby avoiding transmission of incompatible models and reducing resource overhead while ensuring compatibility.
2Reliability
If AI models are transmitted without feedback mechanism, then transmission process is simpler, but model execution status cannot be verified and compatibility cannot be ensured
Solution Approach 1:
The patent implements feedback by requiring the receiver to report model execution status, compatibility information, and performance metrics back to the transmitter. This feedback loop enables the transmitter to understand which models work correctly and which require adjustment, thereby ensuring reliability while managing complexity through structured information exchange.
3Speed
If AI models are prepared for immediate usage, then response time is faster, but preparation time for model selection and transmission increases
Solution Approach 1:
The patent applies segmentation by dividing the model transmission process into distinct phases: attribute information transmission, capability evaluation, model selection, and actual model transmission. This segmentation allows parallel processing of evaluation and selection while preparing for immediate usage, thereby reducing overall preparation time while ensuring model readiness.
Data Source
AI summary
An example method performed by a user equipment (UE) comprises receiving, from a base station, switching indication information for switching from a source processing approach to a target processing approach, wherein the source processing approach or the target processing approach are related to an artificial intelligence (AI) model. The method comprises applying the target processing approach from a second time. The second time is identified based on at least one of features of the target processing approach, features of the source processing approach, or a first time. The first time is a time in which the switching indication information is received.