MU-MIMO Precoding Using Orthogonal Projection Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing MU-MIMO precoding methods face challenges in efficiently serving multiple users due to high overhead requirements for accurate channel state information and interference issues, particularly when using Zero-Forcing precoders, which are cumbersome to obtain and require complex calculations.
Innovation Solution
The proposed solution employs compressed channel state information in the form of precoder matrix feedback from wireless devices to calculate new precoder matrices that minimize inter-user interference, using orthogonal projection matrices and linear mappings to determine modified precoder matrices, allowing for efficient MU-MIMO transmission with reduced overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If Zero-Forcing precoders are used for MU-MIMO transmission, then inter-user interference is minimized, but the overhead and complexity of obtaining accurate channel state information increases significantly
Solution Approach 1:
The patent extracts only the essential precoder matrix information from the full channel state information, using compressed feedback that contains only the precoder indices rather than complete channel matrices. This reduces the overhead while maintaining the ability to calculate interference-minimizing precoders.
Solution Approach 2:
The patent changes the parameter representation from full channel state information to compressed precoder matrix indices. By transforming the information format and using codebook-based quantization, the system reduces feedback overhead while preserving the key information needed for interference mitigation.
2Measurement precision
If accurate channel state information is obtained through high overhead methods, then precoding accuracy improves, but the system overhead and resource consumption increase
Solution Approach 1:
The patent uses codebook-based precoding where pre-defined precoder matrices are stored in both the network node and wireless device. Instead of transmitting full channel information, only indices pointing to matching codebook entries are fed back, creating a compressed copy of the essential information.
Solution Approach 2:
The patent performs preliminary quantization and codebook matching at the wireless device side before feedback transmission. By pre-processing the channel information to identify the best matching codebook entries, the system reduces the amount of data that needs to be transmitted while maintaining accuracy.
3Productivity
If traditional MU-MIMO precoding methods are used, then multiple users can be served simultaneously, but the calculation complexity increases due to large dimensional matrix inversion
Solution Approach 1:
The patent segments the precoding process into independent per-user operations. Instead of performing a single large-scale matrix inversion for all users, the system calculates precoders for each user independently based on their individual channel information and the codebook structure, dividing the complex problem into manageable parts.
Solution Approach 2:
The patent uses codebook-based quantization that provides sufficient precision for practical purposes without requiring exact full-rank precoders. By accepting a small loss in precision due to codebook quantization, the system avoids the computationally expensive exact matrix inversion while maintaining adequate performance.
Data Source
AI summary
A network node is configured to receive and/or estimate respective precoder information for a plurality of wireless devices. A respective orthogonal projection matrix is determined for each of the plurality of WDs based at least in part on the precoder information of all other WDs of the plurality of WDs. A respective modified precoder matrix is determined for each of the plurality of WDs based at least in part on the respective orthogonal matrix and precoder information for each WD of the plurality of WDs. A composite precoder matrix is determined based at least in part on the modified precoder matrices for each of the plurality of WDs. A transmission beam shape is then determined for use in transmission to at least one of the plurality of WDs based on the composite precoder matrix.