CSI Report Prioritization Using Encoder-Based Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing priority schemes for channel state information (CSI) reports in wireless communication systems, particularly in New Radio (NR) systems, fail to meet the increased demands due to the rise in antenna numbers and stringent performance requirements, leading to excessive overheads and redundancy.
Innovation Solution
A method and apparatus that determine the priority of CSI reports based on an encoder, using artificial intelligence or machine learning, which balances accuracy and overheads, and is applicable to both AI/ML-based and traditional CSI report schemes, reducing hardware complexity and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional PMI feedback mode is used with increased antenna numbers, then system performance requirements can be met, but redundant overheads increase significantly
Solution Approach 1:
The patent extracts and removes redundant information from CSI reports by using AI/ML-based compression techniques. The system identifies and eliminates duplicate or unnecessary feedback elements while retaining only the essential channel state information needed for reliable communication, thereby reducing overhead without compromising system performance.
Solution Approach 2:
The patent changes the parameters of CSI reporting by transitioning from traditional detailed PMI/CQI/RI reports to compressed representations generated by AI/ML models. This parameter transformation reduces the quantity of feedback information while maintaining the critical performance-determining characteristics through intelligent compression algorithms.
2Reliability
If more CSI reports are generated to meet stringent performance requirements, then system performance improves, but overheads and complexity increase
Solution Approach 1:
The patent merges multiple CSI reporting functions into a unified AI/ML-based compression framework. Instead of separately handling PMI, CQI, and RI reports with individual processing chains, the system consolidates these into an integrated model that generates compressed feedback, thereby reducing hardware complexity while meeting performance requirements.
Solution Approach 2:
The patent replaces traditional mechanical CSI processing systems with AI/ML-based computational models. The complex hardware architectures required for processing multiple detailed CSI reports are substituted with software-based intelligent compression algorithms that achieve the same or better performance with reduced hardware complexity.
3Quantity of substance
If traditional CSI compression schemes are used, then overheads are reduced, but they fail to satisfy new needs for channel information feedback
Solution Approach 1:
The patent introduces dynamic adaptability into CSI compression by using AI/ML models that can adjust their compression strategies based on varying channel conditions, traffic types, and performance requirements. This dynamic approach allows the system to satisfy diverse new needs for channel information feedback while maintaining reduced overheads, unlike static traditional compression schemes.
Solution Approach 2:
The patent creates a universal AI/ML-based compression framework that can handle multiple types of CSI feedback needs simultaneously. The same compression architecture adapts to different service requirements, frequency ranges, and channel scenarios, providing versatile support for new feedback needs while maintaining overhead reduction across all applications.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
Disclosed in the present application are a method and apparatus used in a node for wireless communication. A first receiver receives first signaling, wherein the first signaling is used for determining N CSI reports, N being a positive integer, and the N CSI reports comprise a first CSI report; and a first transmitter sends a first CSI set, wherein the first CSI set comprises CSI for the N CSI reports; and the first CSI report is used for determining a first RS resource set, the measurement for the first RS resource set is used for determining first pre-compressed CSI, the first pre-compressed CSI serves as an input of a first encoder and is used for generating CSI for the first CSI report, and the priority of the first CSI report depends on the first encoder. The present application reduces system overheads, and improves the performance.