MIMO CSI Compression Using Multi-Dimensional RF Signature Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional CSI feedback in 3GPP 5G NR MIMO networks lacks comprehensive information for multi-use MIMO transmission scheme determination, necessitating enhancements in MIMO channel modeling for improved CSI feedback.
Innovation Solution
User equipment estimates a multi-dimensional non-orthogonal basis matrix and coefficient matrix from CSI-RSs, compresses these matrices, and feeds them back to the network with specific periodicities, employing techniques like eigenvalue or singular value decomposition for further reduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional CSI feedback is used to provide precoder information, then the feedback overhead is reduced, but the information completeness for MIMO transmission scheme determination is insufficient
Solution Approach 1:
The patent segments the downlink channel matrix into two components: a basis matrix representing spatial domain characteristics and a coefficient matrix representing linear combination weights. This segmentation allows selective feedback of only essential components, providing complete information for MIMO transmission while reducing feedback overhead by excluding redundant data.
Solution Approach 2:
The patent extracts and feeds back only the critical components (basis matrix and coefficient matrix) necessary for MIMO transmission scheme determination, rather than feedback the entire channel matrix. This extraction approach maintains information completeness for transmission decisions while significantly reducing feedback overhead.
2Loss of information
If the basis matrix and coefficient matrix are fed back without compression, then the information completeness is maintained, but the feedback overhead increases
Solution Approach 1:
The patent applies parameter changes by using eigenvalue decomposition or singular value decomposition to transform the basis matrix and coefficient matrix into compressed representations. These mathematical transformations reduce the number of parameters that need to be fed back while preserving the essential information needed for accurate channel state representation and MIMO transmission optimization.
3Measurement precision
If full channel matrix information is provided for MIMO transmission, then the transmission accuracy is improved, but the feedback complexity increases
Solution Approach 1:
The patent segments the channel matrix into basis and coefficient components, where the basis matrix captures the dominant spatial characteristics and the coefficient matrix provides linear combination weights. This segmentation maintains measurement precision by preserving essential channel information while reducing feedback complexity through selective reporting of only the most significant components.
Solution Approach 2:
The patent transforms the channel representation using eigenvalue or singular value decomposition, changing the parameters from the original channel matrix elements to decomposed matrix components. This parameter transformation maintains channel state accuracy by preserving the essential spectral and spatial characteristics while significantly reducing feedback complexity through dimensionality reduction.
Data Source
AI summary
Method and user equipment (UE) are provided for CSI compression based on multi-dimensional MIO RF signature. In one novel aspect, the UE receives CSI-RS, estimates a basis matrix and a coefficient matrix of a downlink channel matrix based on the at least one CSI-RS, wherein the basis matrix is an N-dimensional non-orthogonal matrix, with N greater than two, and the coefficient matrix is a linear combination coefficient matrix for the basis matrix, and transmits to the network at least one feedback comprising feeding back the basis matrix in a first periodicity and feeding back the coefficient matrix in a second periodicity. In one embodiment, the UE compresses the basis matrix and the coefficient matrix to a feedback basis matrix and a feedback coefficient matrix before transmitting.