Channel-Aware CSI Feedback Compression for Overhead-Performance Balance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI or ML based CSI feedback methods lack details for effective overhead reduction and performance balancing, particularly in adapting compression methods based on channel quality information.
Innovation Solution
A terminal device determines CSI and channel quality information, compresses CSI using a compression method based on this information, and transmits it to a network device, while the network device recovers the CSI using the same method, allowing for dynamic compression based on channel quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If codebook based CSI feedback is used, then CSI feedback overhead is reduced, but compression performance and adaptability to channel quality are insufficient
Solution Approach 1:
The patent implements dynamic compression methods that adapt to channel quality conditions. The compression algorithm adjusts its parameters based on real-time channel quality information, transitioning between different compression levels or methods depending on the channel state. This dynamic adaptation resolves the contradiction by making the compression performance versatile across different channel conditions while maintaining overhead reduction benefits.
Solution Approach 2:
The patent changes compression parameters based on channel quality measurements. When channel quality is good, lower compression ratios are used to preserve CSI accuracy. When channel quality degrades, higher compression ratios are applied to maintain overhead efficiency. This parameter adaptation allows the system to optimize both overhead reduction and compression performance according to actual channel conditions.
2Loss of information
If AI or ML based compression method is applied, then CSI feedback overhead is significantly reduced, but details for effective overhead reduction and performance balancing are incomplete
Solution Approach 1:
The patent introduces channel quality information as an intermediary that bridges the gap between AI/ML compression methods and practical implementation. This intermediary parameter enables the complex AI/ML-based compression to be controlled and adapted based on measurable channel conditions, providing a practical mechanism for performance balancing without requiring complete detailed specifications for all possible scenarios.
Solution Approach 2:
The patent performs preliminary channel quality assessment before applying AI/ML based compression. By evaluating channel conditions in advance, the system can select appropriate compression strategies and parameters, simplifying the implementation process and providing clear guidance for overhead reduction without needing to handle all complexity in real-time.
3Productivity
If dynamic compression method based on channel quality is applied, then communication efficiency is improved, but system complexity increases
Solution Approach 1:
The patent segments the compression method determination into distinct steps: channel quality measurement, compression method selection based on quality thresholds, and application of the selected method. This segmentation reduces system complexity by breaking down the complex decision-making process into manageable, independent stages that can be implemented and controlled separately while still achieving improved communication efficiency.
Data Source
AI summary
Embodiments of the present disclosure relate to methods, devices and computer readable media for communication. A terminal device determines CSI and channel quality information based on a measurement on a set of RSs from a network device, and determines a compression method at least based on the channel quality information. The terminal device compresses the CSI based on the compression method, and transmits, to the network device, the channel quality information and the compressed CSI. The network device determines the compression method applied for the compressed CSI based on the channel quality information, and recoveries CSI based on the compressed CSI and the compression method. In this way, CSI feedback overhead and compression performance may be balanced.