Beam Prediction Model Feedback for Stable AI Performance Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The performance of AI models used for beam prediction in mobile communication systems is unstable and lacks accuracy due to varying conditions, leading to inconsistent beam quality measurements.
Innovation Solution
A method and apparatus for sending and receiving a performance indication to indicate that the AI model's performance is above a threshold, allowing for improved beam prediction accuracy by ensuring the network device uses a reliable model for subsequent predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If AI model is used for beam prediction, then measurement amount is reduced, but model performance stability deteriorates
Solution Approach 1:
The terminal feeds back performance indication information to the network device, providing feedback on the actual performance of the AI model. This feedback mechanism allows the network device to adjust or switch models based on actual performance, resolving the contradiction by maintaining productivity while improving reliability through continuous performance monitoring and adaptation.
Solution Approach 2:
The system dynamically switches between different AI models or between AI model and traditional measurement methods based on performance indications. This dynamic adaptation allows the system to maintain high productivity when AI models perform well while ensuring reliability by switching to alternative methods when performance degrades, thus resolving the stability issue.
2Measurement precision
If AI model performance validation is added, then beam prediction accuracy is improved, but system complexity increases
Solution Approach 1:
The terminal autonomously evaluates the performance of the AI model and generates performance indication information without requiring complex external validation systems. This self-service approach improves beam prediction accuracy through validation while minimizing system complexity by leveraging existing terminal capabilities rather than adding separate validation infrastructure.
Data Source
AI summary
A method for sending a performance indication, performed by a terminal, includes: sending the performance indication to a network device, wherein the performance indication is configured to indicate that a performance of a model is higher than a threshold, and the model is used for beam prediction.