Wireless Node CSI Reporting Timing for AI-Based Feedback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The integration of ML/AI-based CSI reporting in wireless communication systems requires different processing capabilities and unit occupation times compared to traditional CSI reporting, leading to inefficiencies and potential waste of processing resources.

Innovation Solution

A method and apparatus that dynamically adjust the processing unit occupation time based on the type of CSI reporting, whether it is ML/AI-based or traditional, ensuring optimal allocation and utilization of processing resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If ML/AI-based CSI reporting is introduced to enhance CSI feedback accuracy, then CSI feedback accuracy is improved, but processing unit occupation time increases

Engineering Contradiction:
ImproveCSI feedback accuracyVSAvoidprocessing unit occupation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies dynamics by making the processing unit occupation time configurable and adaptable based on the CSI reporting type. The system dynamically adjusts the time resources allocated for processing between traditional and ML/AI-based CSI reporting methods, allowing the processing duration to be optimized according to whether machine learning techniques are employed, thereby resolving the contradiction between improved accuracy and increased time consumption.

Inventive Principle:
Principle #15Dynamics

2Device complexity

If a unified processing time is allocated for all CSI reporting types, then system design is simplified, but processing resources are wasted

Engineering Contradiction:
Improvesystem design complexityVSAvoidprocessing resource waste
Core Design Contradiction:
Device complexityVSLoss of energy

Solution Approach 1:

The patent implements local quality by allocating different processing time resources to different CSI reporting types based on their specific needs. Instead of applying a uniform processing time across all CSI reports, the system tailors the processing duration to match the computational requirements of each reporting type (traditional vs. ML/AI-based), thereby avoiding waste of processing resources while maintaining manageable system design through structured time allocation.

Inventive Principle:
Principle #3Local quality

3Reliability

If processing unit occupation time is extended to accommodate ML/AI calculations, then calculation completion is ensured, but system productivity decreases

Engineering Contradiction:
Improvecalculation completion reliabilityVSAvoidsystem productivity
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies periodic action by implementing a structured time allocation mechanism where processing units are assigned to ML/AI-based CSI reporting during specific time periods and to traditional CSI reporting during other periods. This periodic scheduling ensures that ML/AI calculations receive sufficient time to complete reliably while preventing continuous occupation of processing units, thereby maintaining system productivity through organized temporal distribution of computational tasks.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentEP4693966A1Method and apparatus used in node for wireless communication
Publication Date: 2026.02.11 APOGEE 5G GLOBAL LLC
  • EP4693966A1 patent drawingFigure 1~2
  • EP4693966A1 patent drawingFigure 3
  • EP4693966A1 patent drawingFigure 4

AI summary

Disclosed in the present application are a method and apparatus used in a node for wireless communication. The method comprises: a first node receiving first signaling; receiving a first RS in a first RS resource; and sending first channel information, wherein the starting of a first symbol which carries the first channel information is not earlier than a Y-th symbol after the last symbol which is occupied by the first signaling; the value of Y is related to whether the first channel information comprises a first type, and the first type meets at least one of the following conditions: channel information which is recovered by a receiver on the basis of the received first type is unknown to the first node; a generator for generating the first type is obtained on the basis of training; and the first type does not belong to CSI which is defined by 3GPP Rel-17 or a previous version. The present application improves the processing of CSI reporting in different scenarios, and optimizes the occupation time of a processing unit according to requirements, thereby avoiding the waste of the processing capability while meeting the requirements of CSI reporting for the processing capability.