Channel Measurement Precoding Across Subbands for Accurate CSI Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing channel measurement methods in Massive MIMO systems suffer from quantization errors due to the discrete and finite codebook sets not matching the continuous and infinite real channels, limiting the precision of downlink channel state information (CSI) feedback.
Innovation Solution
The method involves precoding reference signals with different matrices in multiple frequency domain units within the same time domain unit, reducing correlation between channels and quantization feedback errors, and combining codebook feedbacks across time-frequency blocks to improve downlink channel reconstruction precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a discrete and finite codebook set is used for channel measurement, then feedback overhead is reduced, but quantization error increases and measurement precision deteriorates
Solution Approach 1:
The patent divides the frequency domain into multiple first frequency domain units (e.g., subbands) and applies different precoding matrices to reference signals in each unit. This segmentation allows the terminal device to perform codebook quantization independently on each unit, reducing the correlation between quantization errors across different frequency units and thereby improving overall measurement precision without increasing feedback overhead.
Solution Approach 2:
The patent applies different precoding matrices corresponding to different first frequency domain units when sending reference signals. This local differentiation ensures that each frequency unit has optimized precoding tailored to its specific channel characteristics, reducing the correlation between channels in different frequency units and improving the precision of downlink channel state information feedback.
2Device complexity
If the same precoding matrix is used for reference signals in all frequency domain units, then device complexity is reduced, but correlation between channels in different frequency units increases
Solution Approach 1:
The patent assigns different precoding matrices to reference signals in different first frequency domain units. This local differentiation reduces the correlation between channels in different frequency units, improving the reliability of channel state information feedback while maintaining manageable device complexity through structured precoding matrix selection.
Data Source
AI summary
A channel measurement method includes: receiving a plurality of reference signals in a plurality of first frequency domain units in a same time domain unit, where the plurality of reference signals may be respectively precoded by using precoding matrices corresponding to the first frequency domain units in which the plurality of reference signals are located, and precoding matrices corresponding to at least two different first frequency domain units in the plurality of first frequency domain units are different; generating a precoding matrix indicator (PMI) based on the plurality of reference signals, where the PMI indicates a plurality of codewords corresponding to a plurality of second frequency domain units, the plurality of second frequency domain units and the plurality of first frequency domain units belong to a same frequency domain resource, and the plurality of codewords may be used for determining a downlink channel; and sending the PMI.