Multi-Resolution PMI Feedback for MU-MIMO Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MU-MIMO scheduling efficiency methods in wireless networks face challenges due to imperfect SU-MIMO PMI feedback, leading to suboptimal user pairing and rate allocation, especially with quantized PMI feedback causing spatial multi-user interference.
Innovation Solution
The proposed solution enhances PMI and MU-CQI feedback by computing and sending projections of error vectors along the best directions, allowing the eNB to improve ZF precoding and rate allocation, using Multi-Resolution Desired PMI and Multi-Resolution Companion PMI approaches to provide more accurate direction feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If quantized PMI feedback is used, then feedback overhead is reduced, but spatial multi-user interference increases and scheduling efficiency decreases
Solution Approach 1:
The patent segments the PMI feedback into two components: a first PMI indicating a first precoding matrix from a first codebook, and a second PMI indicating a second precoding matrix from a second codebook. This segmentation allows the UE to provide more precise directional information while controlling feedback overhead by utilizing codebook subsets with specific orthogonality properties.
Solution Approach 2:
The patent changes the parameter of codebook selection by restricting the second codebook to contain only precoding matrices that are (quasi-)orthogonal to the first precoding matrix. This parameter change enables the system to achieve better user orthogonality and reduce spatial multi-user interference while maintaining manageable feedback overhead through structured codebook design.
2Measurement precision
If multiple companion PMIs are evaluated, then user pairing accuracy is improved, but implementation complexity increases
Solution Approach 1:
The patent applies local quality by evaluating multiple companion PMIs only for specific users who are co-scheduled with the target user. Instead of evaluating all possible users, the UE focuses the evaluation on a limited set of co-scheduled users, reducing computational complexity while maintaining pairing accuracy for the relevant user group.
Solution Approach 2:
The patent uses partial action by having the UE evaluate MU-CQI for only K>1 companion PMIs rather than all possible PMIs. The UE draws companion PMI hypotheses from a codebook subset with (quasi-)orthogonal column spaces, providing sufficient evaluation coverage for practical purposes while avoiding excessive computational burden.
3Object-affected harmful factors
If codebook subset restriction is applied, then spatial multi-user interference is reduced, but feedback precision is degraded
Solution Approach 1:
The patent introduces a second dimension of codebook selection beyond the traditional single PMI approach. By selecting a first PMI from a first codebook and a second PMI from a second codebook of (quasi-)orthogonal matrices, the system adds a dimensional layer to the feedback that enables both interference reduction through orthogonality and precision through multi-component specification.
Data Source
AI summary
Disclosed are methods and a device for Multi-resolution PMI Feedback. In one implementation, a user equipment finds a rank 1 or rank 2 Precoding Matrix Indicator based on the signal channel matrix and interference covariance matrix, defines an error vector, obtains an orthonormal basis for the projection matrix, finds the (M−1)-dimensional vector from a codebook (e.g., oversampled Discrete Fourier Transform) with the minimum Euclidean distance, and sends a feedback representing to the base station regarding the vector that it found in the codebook.