Reducing MIMO CSI Computation via DFT Vector Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current massive MIMO wireless communication systems face high computational complexity in deriving Precoding Matrix Indicator (PMI) and Rank Indicator (RI), which hinders efficient resource scheduling and spatial multiplexing, especially in 5G NR systems with large numbers of antennas.
Innovation Solution
The method involves determining correlation values between Discrete Fourier Transform (DFT) vectors and channel observations, selecting DFT vectors with correlation values above a threshold, and using this subset to derive CSI, thereby reducing the computational complexity of PMI and RI determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the number of antennas in massive MIMO systems is increased to improve data rate and link reliability, then the computational complexity of deriving PMI and RI increases significantly
Solution Approach 1:
The codebook is segmented into multiple groups, each corresponding to different antenna subsets. Instead of evaluating all antennas simultaneously, the system divides the large-scale antenna array into smaller manageable groups, processes each group separately to derive PMI and RI, and then combines results. This segmentation reduces the computational burden while maintaining the benefits of massive MIMO.
Solution Approach 2:
The patent applies partial action by evaluating only a subset of DFT vectors rather than all possible vectors in the codebook. By selecting and processing only the most relevant vectors based on channel conditions, the system achieves sufficient accuracy for PMI and RI derivation without the excessive computational cost of exhaustive search, thus resolving the contradiction between accuracy and complexity.
2Measurement precision
If exhaustive search of all DFT vectors is performed to ensure accurate CSI derivation, then measurement precision is improved but processing time increases
Solution Approach 1:
The system performs preliminary actions by pre-organizing the codebook into structured groups and pre-identifying relevant DFT vectors based on channel statistics before actual PMI and RI derivation. This preliminary organization allows the system to skip unnecessary evaluations during real-time operation, maintaining CSI accuracy while significantly reducing processing time.
Solution Approach 2:
The patent extracts and focuses only on the most relevant DFT vectors that contribute significantly to channel representation. By taking out and processing only these critical vectors rather than all vectors in the codebook, the system maintains measurement precision for CSI derivation while reducing the number of operations required, thus decreasing processing time.
Data Source
AI summary
Provided is a method of determining Channel State Information (CSI) in a multiple input/multiple output (MIMO) wireless communication system. The CSI may comprise a Precoding Matrix Indicator (PMI) and/or a Rank Indicator (RI). The method comprises, for a matrix of channels comprising a link between a gNodeB (gNB) and a user equipment (UE), determining correlation values between all Discrete Fourier Transform (DFT) vectors and the observations from the channel matrix. The DFT vectors may include the horizontal vector direction and the vertical vector direction. The method includes selecting those DFT vectors in one or more selected vector directions having a correlation value greater than a predefined threshold to thereby identify a subset of all DFT vectors and determining said CSI from the selected subset of DFT vectors.