Channel State Information Feedback Overhead Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Frequency Division Duplexing (FDD) systems, the high-precision codebook structure in MIMO systems requires terminal devices to feed back a large quantity of precoding matrix indicator bits, leading to increased feedback overhead and reduced resource utilization.
Innovation Solution
The proposed method reduces feedback overhead by selectively sending precoding indication information for only a subset of frequency domain subbands, using interpolation to determine the remaining elements, thereby improving feedback precision and resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a high-precision codebook structure is introduced in MIMO system, then feedback precision is improved, but feedback overhead is increased
Solution Approach 1:
The codebook is segmented into multiple subcodebooks, where each subcodebook corresponds to a specific subband. Instead of selecting one precoding matrix from a single large codebook covering all subbands, the terminal device selects precoding matrices from different subcodebooks for different subbands. This segmentation reduces the number of bits required to indicate the precoding matrix while maintaining precision across different frequency domains.
Solution Approach 2:
Different subcodebooks are designed with different local qualities or characteristics suited for different subbands. Each subcodebook is optimized for specific frequency ranges, allowing the system to achieve high feedback precision locally in each subband while reducing overall feedback overhead through selective indication.
2Measurement precision
If precoding matrix indicator bits are increased for high-precision feedback, then feedback precision is improved, but resource utilization is reduced
Solution Approach 1:
By segmenting the feedback into subband-specific precoding matrix indicators, the system reduces the total number of bits required compared to a single large codebook approach. This segmentation allows more efficient use of feedback resources while maintaining the precision needed for accurate channel state information.
Solution Approach 2:
The system uses partial action by indicating precoding matrices only for necessary subbands using a subset of subcodebooks, rather than providing full detailed feedback for all possible precoding matrices across all subbands. This partial indication approach reduces feedback overhead while maintaining sufficient precision for effective precoding.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6a
AI summary
This application provides a channel state information feedback method and an apparatus. The method includes: generating first frequency domain indication information and M1 pieces of first precoding indication information, where the first frequency domain indication information is used to indicate L1 frequency domain subbands in T frequency domain subbands, the T frequency domain subbands are a system bandwidth or a part of the system bandwidth, 1 ≤ L1 < T, the T frequency domain subbands are in a one-to-one correspondence with T precoding matrices, and a precoding matrix wk corresponding to a kth frequency domain subband satisfies: Wk=W1×W2k, where w1 represents a matrix with Nt rows and R columns, and W2k represents a matrix with R rows and S columns; and w1 represents a matrix with Nt rows and R columns, and W2k represents a matrix with R rows and S columns; and the first frequency domain indication information and the M1 pieces of first precoding indication information are used to determine T first elements, a kth first element in the T first elements is a factor of an element in an rth row and an sth column of W2k, and 1 < M1 < T; and sending the first frequency domain indication information and the M1 pieces of first precoding indication information. In the method, feedback overheads of channel state information are reduced.