MU-MIMO Scheduling Using Long-Term CSI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high feedback overhead in wireless communication systems due to the need for full channel state information (CSI) in MU-MIMO systems, especially in large user scenarios, restricts the practical application of MU-MIMO, as existing solutions like codebooks do not adequately reduce the overhead.
Innovation Solution
Employing long-term statistical CSI, specifically mean channel matrices, for user scheduling, while maintaining instantaneous CSI for precoding, to reduce feedback overhead and simplify the scheduling process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full instantaneous CSI is used for all users in MU-MIMO scheduling, then scheduling performance is improved, but feedback overhead becomes prohibitive
Solution Approach 1:
The patent segments the CSI feedback process into two distinct phases: a scheduling phase that uses long-term statistical CSI (mean channel matrices) and a precoding phase that uses instantaneous CSI. This segmentation allows the system to reduce feedback overhead during scheduling while maintaining performance through the use of statistical CSI, which captures essential channel characteristics without requiring full instantaneous details for all users.
Solution Approach 2:
The patent changes the parameter used for scheduling from instantaneous CSI to long-term statistical CSI (mean channel matrices). This parameter change fundamentally alters the feedback requirements, reducing the overhead from reporting full instantaneous channel states to reporting only statistical characteristics that change slowly over time, thereby resolving the contradiction between scheduling performance and feedback overhead.
2Loss of information
If codebook technique is used to reduce feedback overhead, then per-user feedback is reduced, but overall feedback overhead still grows linearly with total user number
Solution Approach 1:
The patent applies partial action by using long-term statistical CSI instead of full instantaneous CSI for scheduling. This partial approach provides sufficient information for effective scheduling while significantly reducing the feedback overhead, avoiding the need to report complete instantaneous channel states for all users and thereby reducing the linear growth of feedback overhead with the number of users.
3Loss of information
If long-term statistical CSI is used for scheduling, then feedback overhead is reduced, but scheduling accuracy may be compromised
Solution Approach 1:
The patent introduces dynamics by using long-term statistical CSI that changes slowly over time rather than instantaneous CSI that changes rapidly. This dynamic approach allows the system to update CSI at a lower rate, reducing feedback overhead while maintaining sufficient accuracy for scheduling decisions. The statistical CSI captures the essential temporal characteristics of the channel, providing a balance between overhead reduction and scheduling accuracy.
Data Source
AI summary
The present invention proposes a multiuser multi-input multi-output scheduling method, which comprises steps of: a multiuser scheduling step of performing multiuser scheduling by using mean channel matrixes of respective user equipments; and a multiuser precoding step of performing multiuser precoding by using instantaneous channel matrixes of respective user equipments selected in the multiuser scheduling. The present invention also proposes a base station and a user equipment which will be used to implement the inventive multiuser multi-input multi-output scheduling method.