Implicit CSI Feedback for DL MU-MIMO Link Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing DL MU-MIMO systems, such as E-UTRA LTE Rel-8, face limitations due to unreliable Channel Quality Indicator (CQI) feedback, which is based on single-user MIMO precoding, leading to inefficient link adaptation and scheduling in multi-user environments.
Innovation Solution
The implementation of implicit CSI feedback, where user equipment reports multiple single-user precoding matrix indicators to facilitate beamforming and improve channel state information reporting, allowing the base station to derive optimal precoding matrices for multi-user MIMO operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If single-user MIMO precoding is used for CQI feedback, then the feedback mechanism is simple and compatible with existing systems, but the accuracy and reliability of CQI feedback deteriorates in multi-user environments
Solution Approach 1:
The precoding matrix indicator feedback is segmented into multiple components: a first PMI indicating recommended optimum precoding matrix and further PMIs indicating information for beamforming at the base station. This segmentation allows the feedback to capture both optimal precoding information and beamforming directions, improving accuracy while maintaining compatibility with existing single-user feedback mechanisms.
Solution Approach 2:
The patent extends the traditional single-dimension CQI feedback by adding multiple PMI dimensions. Instead of reporting only a single precoding matrix indicator, the system now reports multiple PMIs that provide information about optimal precoding matrices and beamforming directions, creating a multi-dimensional feedback structure that better characterizes the channel state for MU-MIMO operations.
2Adaptability or versatility
If codebook-based precoding is used for MU-MIMO, then the precoding matrices can be selected from a pre-defined set, but the system cannot account for actual interference received on data transmission
Solution Approach 1:
The patent implements enhanced feedback mechanisms where user equipment reports multiple PMIs that include information about beamforming directions and optimal precoding matrices. This feedback allows the base station to infer actual interference conditions and adjust link adaptation parameters accordingly, moving beyond the limitations of pre-defined codebook selection alone.
Solution Approach 2:
The system performs preliminary beamforming based on the reported PMIs before actual data transmission. By using the first PMI to determine beamforming directions and the further PMIs to account for interference conditions, the base station can pre-configure precoding matrices that optimize performance for the specific multi-user scenario before transmission begins.
3Ease of operation
If CQI feedback is based on single user MIMO precoding, then the feedback mechanism is straightforward, but the performance of MU-MIMO systems is limited
Solution Approach 1:
The patent introduces dynamic precoding matrix selection by reporting multiple PMIs that can adapt to changing channel conditions and user configurations. The first PMI dynamically indicates the recommended optimum precoding matrix while further PMIs dynamically provide beamforming information, allowing the system to optimize MU-MIMO performance in real-time while maintaining a feedback structure similar to existing single-user mechanisms.
Data Source
AI summary
This invention is method of implicit channel state information (CSI) feedback for multiuser multiple input, multiple output (MU-MIMO) communication between a base station (eNB) and a plurality of user equipment (UE). Each user equipment reporting a plurality of single user precoding matrix indicators (PMI) including a first PMI indicating a recommended optimum precoding matrix and at least one further PMI indicating information for beamforming at said base station. The base station communicates with each user equipment based on precoding matrix indicators received from the plurality of user equipment.