UE Anchor-Based Data Reporting for AI Wireless Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
AI-based wireless communication systems face issues with low generalization due to poor data quality and high air interface overhead, leading to biased and unreliable AI models, especially when user equipment (UE) data is not adequately evaluated for quality before reporting.
Innovation Solution
User equipment (UE) determines data quality by comparing it to configured anchors, reporting high-quality data to the base station (BS) using predefined thresholds and priorities, ensuring accurate and efficient training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If UE reports all available data to the base station, then the base station can collect more training data, but the air interface overhead increases significantly
Solution Approach 1:
The patent extracts only the essential quality indicator (difference value) from the complete data set. Instead of transmitting all raw data, the UE calculates a compact difference value that represents data quality relative to anchors, achieving selective extraction of critical information that reduces overhead while preserving training value.
Solution Approach 2:
The patent transforms the data representation from raw high-dimensional data to a scalar difference value parameter. This parameter transformation compresses the information while maintaining the essential quality characteristic needed for training, effectively converting complex data into a compact quality metric.
2Ease of operation
If UE reports data without quality evaluation, then reporting is simple, but the AI model training becomes inefficient and produces biased models
Solution Approach 1:
The UE performs self-evaluation of its data quality by autonomously calculating the difference value between its data and the anchors. This self-service quality assessment enables the UE to independently determine whether its data meets reporting criteria, eliminating the need for complex base station-side quality verification while ensuring reliable data submission.
Solution Approach 2:
The patent introduces a feedback mechanism where the UE compares its data against pre-configured anchors and uses the resulting difference value to gate reporting decisions. This feedback loop ensures that only data meeting quality thresholds is reported, improving training reliability while maintaining operational simplicity through automated quality control.
3Quantity of substance
If the base station collects low quality data, then more data samples are available for training, but the training latency increases and model performance deteriorates
Solution Approach 1:
The patent applies preliminary quality filtering at the UE side before data transmission. By pre-calculating difference values and comparing against thresholds, the system performs advance quality assessment that prevents low-value data from entering the training pipeline, ensuring that collected samples are inherently high-quality and ready for immediate training use.
Data Source
AI summary
Embodiments of the present application provide a communication method and a communication apparatus. The method includes: sending first data to a network device when a first difference value is less than or equal to a first threshold value, the first difference value being a difference value between the first data and a first anchor, the first anchor including one or multiple pieces of reference data, the first threshold value being a predefined or configured threshold value corresponding to the first anchor; and performing communication based on the first data.