Differential Quantization of Precoder Matrix Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication systems, particularly in 5G NR, face challenges in efficiently determining and quantizing channel state information for precoder matrix selection and combination, leading to suboptimal beamforming and interference suppression in multi-user MIMO scenarios.
Innovation Solution
The method involves a user equipment (UE) measuring reference signals to determine a channel state with multiple delay taps, calculating a first precoder matrix for beam selection and a second precoder matrix for beam combination, with coefficients normalized and quantized for efficient transmission, employing differential quantization techniques to reduce feedback overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional quantization methods are used for precoder matrix coefficients, then feedback overhead is high, but manufacturing precision (quantization accuracy) is maintained
Solution Approach 1:
The patent applies differential quantization by changing the quantization parameter representation from absolute values to differential values relative to a reference coefficient. Each coefficient is quantized as a differential value from its reference coefficient, reducing the number of bits needed while maintaining accuracy through the relationship: original_coefficient = reference_coefficient + differential_quantized_value
Solution Approach 2:
The patent performs preliminary selection of reference coefficients from the precoder matrix before quantization. By identifying and separating reference coefficients (which are fully quantized) from differential coefficients (which are differential-quantized), the system prepares the data structure in advance to enable efficient differential quantization and reduce feedback overhead
2Measurement precision
If high-resolution channel state information is transmitted, then beamforming accuracy is improved, but feedback overhead increases
Solution Approach 1:
The patent transforms the parameter representation of channel state information by using differential quantization. Instead of transmitting full high-resolution coefficients, the system transmits differential values that represent deviations from reference coefficients, thereby maintaining beamforming accuracy while significantly reducing the feedback overhead for high-resolution channel state information
Solution Approach 2:
The patent segments the precoder matrix coefficients into reference coefficients and differential coefficients. Reference coefficients are fully quantized and serve as benchmarks, while other coefficients are represented as differential values relative to their references. This segmentation allows high-resolution representation where needed while reducing overall feedback complexity
Data Source
AI summary
In an aspect of the disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus may be a UE. The UE measures one or more reference signals to determine a channel state on a channel including M delay taps. The UE determines a first precoder matrix for selecting a set of beams from L beams transmitted from the base station on the channel. The UE determines a second precoder matrix for combining the set of beams. The second precoder matrix includes at least two groups of coefficients each including coefficients being normalized based on a respective reference coefficient. Each coefficient corresponds to a respective one beam of the L beams at a particular polarization on a respective one delay tap of the M delay taps. The UE quantizes coefficients of the second precoder. The UE transmits an indication of the second precoder matrix having quantized coefficients.