MIMO Port Sorting via EVD for Low-Complexity Beamforming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The computational complexity of beam weight calculations in Multiple-Input-Multiple-Output (MIMO) communication systems is high due to the need to use all channel estimates from available SRS ports, leading to inefficient and complex MMSE computations, especially when the number of transmission layers is below the number of antennas.
Innovation Solution
A method involving channel matrix division into sub-matrices, followed by eigen value decomposition (EVD) and port sorting to reduce complexity, where highly correlated channel vectors are grouped and reordered, and port sorting matrices are applied to obtain a final port-sorted channel matrix, reducing computational burden while maintaining performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all channel estimates from available SRS ports are used in beam weight calculations, then beamforming performance is maintained, but computational complexity increases significantly
Solution Approach 1:
The patent segments the channel matrix H into multiple sub-matrices based on SRS port groups. Instead of performing MMSE computation on the complete channel matrix which involves all SRS ports, the system divides it into smaller sub-matrices corresponding to different port groups. This segmentation reduces the computational burden from O(N^3) to multiple smaller O(n^3) operations where n < N, while maintaining beamforming performance through selective port combination.
Solution Approach 2:
The patent extracts and identifies a subset of SRS ports that contribute most significantly to beamforming performance. By performing eigenvalue decomposition or singular value decomposition on the channel covariance matrix, the system extracts the dominant eigenvectors corresponding to the most important SRS ports. This extraction allows the system to use only the essential port information for beam weight calculation, eliminating redundant computations from less significant ports.
2Productivity
If the number of transmission layers is reduced below the number of antennas, then spectral efficiency improves, but port selection complexity increases
Solution Approach 1:
The patent performs preliminary port selection and channel matrix decomposition before the actual beamforming operation. By pre-processing the channel matrix to identify and select the most relevant SRS ports based on channel conditions, the system prepares a reduced-dimensional representation that simplifies subsequent beam weight calculations. This preliminary action ensures that when fewer transmission layers are used, the port selection has already been optimized to match the reduced layer configuration.
Solution Approach 2:
The patent changes the parameter of port grouping by dynamically adjusting how SRS ports are segmented into sub-matrices based on the number of transmission layers. When the number of layers is reduced, the system modifies the port selection parameters to select only the most relevant ports for the reduced layer configuration. This parameter adaptation allows the system to maintain optimal performance with fewer layers while reducing the complexity of port selection through conditional parameter adjustment.
Data Source
AI summary
Systems and methods are disclosed for port selection for a wireless communication system. In one embodiment, a method performed by a Radio Access Network (RAN) node comprises dividing a channel matrix of one subcarrier of a Multiple-Input-Multiple-Output (MIMO) channel between an antenna array of the RAN node and a particular User Equipment (UE) into sub-matrices. The method further comprises forming a re-ordered channel matrix as a concatenation of the sub-matrices and forming a port-sorting matrix based on eigen vector matrices and eigen value matrices obtained via Eigen Value Decompositions (EVDs) performed on channel covariance matrices for the sub-matrices. The method further comprises applying the port-sorting matrix to the re-ordered channel matrix, re-ordering column vectors in the port-sorted channel matrix based on eigen values obtained from the EVDs, and applying grouping of the re-ordered, port-sorted channel matrix and port sorting accordingly to obtain a final port-sorted channel matrix.