MCS Adaptation Complexity Reduction via QR Decomposition Reuse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional modulation coding scheme (MCS) adaptation for multi-user multiple input multiple output (MU-MIMO) downlink scheduling becomes increasingly complex with the addition of users, approaching cubic complexity with respect to the number of transmit antennas, which hampers network resource efficiency.
Innovation Solution
The method involves selecting initial user groups, performing QR decomposition only once for the initial user group, and subsequently calculating new modulation coding schemes for added users directly using information from the initial decomposition, thereby avoiding repeated QR decomposition and matrix inversion operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional MCS adaptation is used for MU-MIMO downlink scheduling, then accurate MCS prediction is achieved, but computational complexity increases to cubic order with respect to the number of transmit antennas
Solution Approach 1:
The patent performs QR decomposition on the channel matrix H once to obtain Q and R matrices before user selection. These pre-computed matrices are then reused for MCS adaptation across multiple user groups, eliminating the need to perform QR decomposition repeatedly for each user group evaluation. This preliminary action reduces computational complexity from cubic to quadratic order while maintaining MCS prediction accuracy.
Solution Approach 2:
The patent makes the channel matrix H (and its QR decomposition results Q and R) universally applicable across all user group evaluations. Instead of computing H-specific decompositions for each user group, the same H decomposition serves multiple purposes: user selection, MCS adaptation, and signal detection. This universal application of the decomposition results significantly reduces computational overhead.
2Productivity
If users are added to the network to improve network capacity, then network throughput increases, but MCS adaptation complexity approaches cubic order
Solution Approach 1:
The channel matrix H is decomposed once using QR decomposition before user grouping. The resulting Q and R matrices are stored and reused for all subsequent MCS adaptations regardless of how many users are added to the network. This preliminary decomposition action enables the system to handle increased user counts and higher network throughput without experiencing cubic complexity growth.
Solution Approach 2:
The patent changes the computational approach from performing full QR decomposition for each user group (cubic complexity) to using pre-computed Q and R matrices with simplified MCS adaptation formulas (quadratic complexity). This parameter change in the computational method allows the system to scale with network size while maintaining manageable complexity levels.
Data Source
AI summary
The computational complexity of MCS adaptation for linear and non-linear MU-MIMO can be reduced by avoiding QR decomposition during subsequent stages of MCS adaptation. For instance, QR decomposition can be avoided in later stages of MCS adaptation by computing an instant upper right triangular matrix (R1) directly from an earlier upper right triangular matrix (R) and an earlier unitary matrix (U), which were obtained during a previous stage of MCS adaptation. As such, the instant upper right triangular matrix (R1) is obtained without performing QR decomposition on an instant Hermitian matrix (H1H), thereby allowing MCS adaptation to be performed for the new user group with less complexity. Additionally, computational complexity of MCS adaptation for linear MU-MIMO can be further reduced by avoiding matrix inversion during subsequent stages of MCS adaptation.