CSI Data Collection for AI-Based Channel Prediction in 5G

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing CSI-RS configurations in 5G do not support data collection for AI-based CSI prediction, limiting the ability to perform AI model training and monitoring.

Innovation Solution

A data collection method and apparatus that allows terminals to autonomously or based on first information to collect CSI data, including AI model identifier, applicable conditions, and CSI-RS configuration, enabling AI-based CSI prediction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If existing CSI-RS configurations in 5G are used, then network compatibility is maintained, but data collection for AI-based CSI prediction is not supported

Engineering Contradiction:

Improvesupport for AI-based CSI prediction data collectionVSAvoidCSI-RS configuration complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the data collection process by introducing separate indication information for AI model training data collection and AI model monitoring data collection. This allows the system to collect different types of CSI data with different configurations, enabling AI-based CSI prediction while maintaining compatibility with existing 5G CSI-RS configurations through modular addition of new functionality.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent makes the CSI-RS configuration universal by designing it to serve multiple purposes: traditional CSI reporting, AI model training data collection, and AI model monitoring data collection. The indication information mechanism allows a single CSI-RS configuration to be flexibly adapted for different AI-related functions without requiring separate dedicated configurations for each function.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Measurement precision

If AI model training data collection is performed, then AI model accuracy improves, but data collection overhead increases

Engineering Contradiction:

ImproveAI model training accuracyVSAvoiddata collection overhead

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent merges AI model training data collection with existing CSI reporting mechanisms. By using the same CSI-RS resources and reporting channels for both traditional CSI and AI training data, the system avoids duplicating infrastructure and reduces overall overhead while still gathering the necessary data for accurate AI model training.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent enables selective data collection where only the necessary CSI parameters for AI model training are collected in addition to traditional CSI reporting. The indication information allows the network to request only specific data sets needed for AI training, avoiding unnecessary data collection and reducing overhead while maintaining sufficient data quality for model accuracy.

Inventive Principle:

Principle #16Partial or excessive action

3Ease of operation

If autonomous data collection by terminal is enabled, then data collection flexibility improves, but control precision from network side decreases

Engineering Contradiction:

Improvedata collection flexibilityVSAvoidnetwork control precision

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent implements a feedback mechanism where the terminal autonomously collects CSI data based on indication information from the network, then reports the collected data back to the network. The network uses this feedback to adjust future data collection requests, optimizing both terminal flexibility and network control precision through iterative coordination based on actual data quality and network needs.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20260025185A1Data collection method and apparatus for ai-based CSI prediction

Publication Date: 2026.01.22 VIVO MOBILE COMM CO LTD

AI summary

This application discloses a data collection method and apparatus for AI-based CSI prediction, and belongs to the field of communication technologies. The method in embodiments of this application includes any one of the following: autonomously performing, by a terminal, CSI data collection to obtain CSI data; or performing, by a terminal, CSI data collection based on first information, to obtain CSI data, where the CSI data is used for AI model training of AI-based CSI prediction; where the first information includes at least one of the following: AI model identifier information, used to indicate an AI model corresponding to the CSI data; applicable condition information, used to indicate a condition or a scenario for data collection; data collection start condition information; data collection continuity condition information; data collection termination condition information; or data collection-related CSI-RS configuration information, used to indicate CSI-RS configuration information of data collection for AI-based CSI prediction.