Wireless CSI Feedback With Joint Channel Estimation Training

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing MIMO wireless communication systems face challenges in accurately performing CSI estimation and feedback due to the limitations of separate channel estimation and CSI feedback models, leading to degraded signal transmission performance.

Innovation Solution

Implement joint training on a channel estimation model and a CSI feedback model, using different resource densities for reference signals to enhance adaptation and improve overall model performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If separate channel estimation and CSI feedback models are used, then device complexity is reduced, but signal transmission performance deteriorates

Engineering Contradiction:
Improvemodel complexityVSAvoidsignal transmission performance
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent combines separate channel estimation and CSI feedback models into a unified joint training framework. The terminal device performs joint training on both models using the same reference signal inputs, enabling the models to share statistical characteristics and improve overall performance. This merging resolves the contradiction by integrating previously separate processing stages into a coordinated system that maintains lower complexity while achieving better transmission performance.

Inventive Principle:
Principle #5Merging (Combining)

2Measurement precision

If high resource density reference signals are used for training, then model performance is improved, but resource consumption increases

Engineering Contradiction:
ImproveCSI feedback accuracyVSAvoidreference signal resources
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent employs different resource density configurations for reference signals during joint training versus actual operation. During joint training, higher resource density reference signals are used to enable accurate model learning and improve CSI feedback accuracy. However, during normal operation, the system switches to lower resource density configurations, thereby achieving good model performance while controlling resource consumption through parameter adjustment at different operational stages.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4641443A1Wireless communication methods, terminal devices and network device
Publication Date: 2025.10.29 GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP LTD
  • EP4641443A1 patent drawingFigure 1
  • EP4641443A1 patent drawingFigure 2
  • EP4641443A1 patent drawingFigure 3~5

AI summary

A wireless communication method, a terminal device, and a network device are provided. The method includes the following. A terminal device performs joint training on a first model and a second model according to first input information and label channel data. The first input information is channel data obtained by receiving a reference signal by the terminal device based on first configuration information. The label channel data is channel data obtained by receiving the reference signal by the terminal device based on second configuration information. A resource density of the reference signal configured by the first configuration information is less than a resource density of the reference signal configured by the second configuration information. The first model is used for channel estimation based on the first input information to obtain first output information. The second model is used to compress and recover second input information to obtain target CSI. The second input information is determined according to the first output information.