Precoding Matrix Construction via Component Vector Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In MIMO systems, the accuracy of the precoding vector for precoding data is relatively low due to the current method of determining the precoding matrix only using direction information of the full channel, leading to inefficient channel measurement and feedback in hybrid CSI-RS feedback mode.
Innovation Solution
A method that includes indicating and constructing a precoding matrix by feeding back both direction information and superimposition coefficients of component vectors in the first-stage CSI feedback mode, improving the accuracy of the precoding vector by setting multiple component vectors to represent the full channel more accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If only direction information of the full channel is used to determine the precoding matrix, then the system complexity is reduced, but the measurement precision of the precoding vector deteriorates
Solution Approach 1:
The channel measurement process is segmented into two stages: full-channel measurement using non-precoded CSI-RS to obtain direction information, and local-channel measurement using precoded CSI-RS to refine the precoding vector. This segmentation allows each stage to focus on specific aspects of channel characterization, improving overall measurement precision without overwhelming system complexity
Solution Approach 2:
The precoding matrix is determined in advance during the first-stage CSI feedback based on full-channel measurement results. This preliminary precoding matrix is then used to precode the CSI-RS for the second-stage measurement, enabling more accurate local-channel measurement while maintaining manageable system complexity through advance preparation
2Measurement precision
If multiple component vectors with superimposition coefficients are fed back in the first-stage CSI feedback mode, then the measurement precision of the precoding vector is improved, but the loss of information in feedback increases
Solution Approach 1:
The feedback information parameters are changed from simple precoding matrix indices to include both component vector indices and their corresponding superimposition coefficients. This parameter expansion enables more precise representation of the channel characteristics while maintaining a structured feedback format that manages information loss through systematic encoding
Data Source
Figure 1.
Figure 2
Figure 3
AI summary
Embodiments of this application disclose a precoding matrix indication method, a precoding matrix determining method, and a device, and relate to the field of communications technologies, to improve accuracy of an obtained precoding vector for precoding data. The precoding matrix indication method includes: receiving N reference signals, where N≥2, and N is an integer; generating first indication information based on the N reference signals, where the first indication information is used to indicate a plurality of component vectors for constructing a precoding matrix and a superposition coefficient of each component vector of each of M precoding vectors for constructing the precoding matrix, and the precoding matrix is used to precode M reference signals, where 1≤M<N, and M is an integer; and sending the first indication information. The technical solution may be applied to a hybrid CSI feedback mode.