Channel State Information Feedback for Large-Scale MIMO Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current closed-loop MIMO systems in LTE do not provide an effective solution for channel state information feedback when the number of downlink transmit antennas exceeds eight, leading to inefficiencies in channel quality information feedback.
Innovation Solution
A method for channel state information feedback that determines a feedback expression manner based on the number of configured antennas, using T CQI values to express CQI of M codewords and one PMI value, where T is less than M, allowing for efficient feedback even with a large number of antennas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the number of downlink transmit antennas is increased beyond eight, then the MIMO system capacity and spectral efficiency are improved, but the channel state information feedback complexity and overhead increase significantly
Solution Approach 1:
The patent segments the feedback of channel quality information by using different numbers of CQI values (T CQI values for M codewords, where T < M) based on the rank value. This segmentation allows the system to adaptively reduce feedback overhead for higher ranks while maintaining necessary feedback precision, directly resolving the contradiction between system capacity and feedback complexity.
Solution Approach 2:
The patent implements dynamic adaptation of feedback parameters by configuring different feedback expression manners based on rank values. The network device configures the number of CQI values to be fed back according to the actual transmission rank, making the feedback system dynamic and adaptable to different MIMO configurations, thus balancing capacity improvement with feedback overhead control.
2Measurement precision
If the number of CQI values fed back is increased to match the number of codewords, then the channel quality information precision is improved, but the feedback overhead increases
Solution Approach 1:
The patent changes the parameter of CQI feedback quantity based on the rank value. For rank 2, one CQI value is fed back; for rank 3, two CQI values are fed back; and so on. This parameter change strategy ensures that feedback precision is maintained at the minimum necessary level for each rank, avoiding excessive feedback overhead while preserving measurement precision where needed.
Solution Approach 2:
The patent applies local quality by differentiating feedback precision requirements for different codewords based on their mapping to layers. Since not all codewords require independent CQI feedback (especially when T < M), the system applies differentiated feedback quality locally to each codeword situation, optimizing the balance between precision and overhead.
3Loss of information
If separate CQI feedback for each codeword is implemented, then the channel quality representation accuracy is improved, but the feedback resource consumption increases
Solution Approach 1:
The patent applies partial action by feeding back T CQI values for M codewords where T < M for certain rank configurations. This partial feedback approach is sufficient to represent channel quality information accurately for the given MIMO configuration, avoiding the excessive resource consumption of feeding back CQI for every single codeword while maintaining adequate representation accuracy.
Data Source
Figure 1~2
Figure 3~3-1
Figure 4~7
AI summary
The present invention discloses a channel state information feedback method and apparatus. User equipment determines a PMI and a CQI, and a feedback expression manner of the PMI and the CQI, where the feedback expression manner includes a first or a second feedback expression manner, and sends the PMI and the CQI to a network device in the determined feedback expression manner of the PMI and the CQI, where the first feedback expression manner includes: using T CQI values to express CQI of M codewords, and using one PMI value to indicate one PMI, where T is less than M, and there is at least one CQI value that can express CQI of at least two codewords; and the second feedback expression manner includes: using M CQI values to express CQI of M codewords, and using one PMI value to indicate one PMI, where a quantity of bits used in one PUCCH by at least one CQI value is less than 3, and a quantity of bits used in one PUCCH by at least one CQI value is greater than 3. The present invention may be applied to a scenario in which an antenna quantity increases.