MIMO Precoding Feedback via Subband Vector Combination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In MIMO wireless communication systems, the limited codebook space results in a significant deviation of selected precoding vectors from ideal vectors, leading to a suboptimal precoding effect, which affects transmission reliability and throughput.
Innovation Solution
A channel information feedback method that generates and sends description information for each subband, including vector and weighting parameters, to select component vectors from a base codebook for weighted combination, thereby enlarging the codebook space and improving precoding accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a precoding vector is selected directly from a codebook, then feedback overhead is reduced, but the precoding vector deviates greatly from the ideal precoding vector, resulting in limited precoding effect
Solution Approach 1:
The precoding matrix is segmented into multiple column vectors, and the feedback mechanism is segmented into two stages: first feedback narrowband precoding matrices for each subband, then feedback wideband precoding matrices that combine these narrowband matrices. This segmentation allows progressive refinement of precoding accuracy while controlling feedback overhead.
Solution Approach 2:
The patent introduces a two-dimensional feedback structure: narrowband dimension (subband-level precoding matrices) and wideband dimension (aggregated precoding matrices). By adding the wideband dimension that combines narrowband matrices through weighted summation, the system achieves higher precoding accuracy without linearly increasing feedback overhead.
2Manufacturing precision
If the ideal precoding vector is directly fed back, then precoding effect is improved, but feedback overhead becomes very high
Solution Approach 1:
Instead of directly feeding back the ideal precoding vector obtained from SVD decomposition, the patent creates approximated precoding matrices by combining codebook vectors through weighted summation. These copied/approximated matrices closely resemble the ideal precoding vectors while requiring significantly less feedback overhead, as only weighting coefficients and codebook indices need to be transmitted.
3Loss of information
If a codebook space is limited, then feedback overhead is reduced, but the selected precoding vector deviates greatly from the ideal precoding vector
Solution Approach 1:
The patent merges multiple narrowband precoding matrices from different subbands into a wideband precoding matrix through weighted summation. This combining operation expands the effective codebook space, allowing the system to represent channel characteristics that no single codebook entry could capture individually, thereby improving adaptability while maintaining reasonable feedback overhead.
Data Source
Figure 1
Figure 2A~2B
Figure 2C~3
AI summary
Embodiments of the present invention provide a channel information feedback method, a channel information determining method, a receive end device, and a transmit end device. The channel information feedback method includes: generating at least one piece of description information for each of a plurality of subbands included in a wideband, where each piece of description information is used to describe a description vector of a channel matrix corresponding to the subband, the description information includes at least one of a vector parameter and a weighting parameter; and sending the at least one piece of description information. The embodiments of the present invention further provide a channel information determining method, a receive end device, and a transmit end device. A precoding effect can be improved when precoding is performed based on the description vector fed back in the technical solutions provided in the embodiments of the present invention.