AI-Assisted Measurement Feedback for NR Channel Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the NR system, traditional channel measurement methods result in low measurement accuracy and high overhead due to the reporting of measured objects only, without information on unmeasured objects, and limited feedback content.
Innovation Solution
Implementing a method for measurement report that involves determining an association between resource sets and AI models, using AI models to generate report information based on these resource sets, and transmitting this information to the network device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional channel measurement and measurement report methods are used, then the measurement overhead is high, but the measurement accuracy is low
Solution Approach 1:
An AI model is introduced as an intermediary between the measurement resources and the report information generation. The AI model processes measurement data from measured resources and generates comprehensive report information that includes both measured and unmeasured object characteristics, thereby improving measurement accuracy without proportionally increasing measurement overhead
Solution Approach 2:
The AI model is pre-trained with channel state information and measurement data to learn patterns and relationships. This preliminary training enables the model to infer unmeasured object information from measured resources, allowing accurate measurement reports without measuring every object individually
2Loss of information
If traditional measurement report methods are used, then the report content is limited to measured resources, but the feedback information is insufficient
Solution Approach 1:
The AI model serves as an intermediary that transforms limited measurement data into comprehensive feedback information. It infers characteristics of unmeasured resources based on measured data and learned patterns, providing the network device with complete channel state information without requiring direct measurement of all resources
Solution Approach 2:
The AI model creates virtual representations (copies) of unmeasured resource characteristics based on patterns learned from measured resources. These inferred copies provide comprehensive feedback information about the entire channel state without requiring physical measurement of every resource, thus reducing information loss while managing report complexity
Data Source
AI summary
Provided in the embodiments of the present disclosure are a measurement feedback method and apparatus, and a storage medium. The method comprises: determining at least one first resource set and at least one AI model, which has an association relationship with the at least one first resource set, wherein the first resource set is a measurement resource set; and determining feedback information on the basis of the at least one first resource set and the at least one AI model, and sending the feedback information to a network device, wherein the feedback information comprises some or all output information of the at least one AI model. In the present disclosure, at least one first resource set, which needs reasoning performed, and at least one AI model, which has an association relationship with the at least one first resource set, are determined, and reasoning is performed by using the at least one first resource set and the at least one AI model, which are associated with each other, so as to obtain feedback information. By means of the AI model, the measurement and management precision are improved, the measurement overheads of measurement resources are reduced, and the measurement period is shortened.