Precoding Vector Quantization for MIMO Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In MIMO wireless communication systems, the uniform quantization bit quantities for combination coefficients of precoding vectors lead to unsatisfactory system performance due to imprecise precoding vectors and high overheads in feedback information.
Innovation Solution
The method involves differentiating the quantization bit quantities for wideband and narrowband amplitude and phase coefficients of component vectors, allocating more bits to coefficients with greater impact on the precoding vector, thereby improving precision while reducing feedback overheads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If uniform quantization bit quantities are used for all combination coefficients, then the feedback overhead is simplified and managed, but the precoding vector precision deteriorates and system performance becomes unsatisfactory
Solution Approach 1:
The patent applies local quality by differentiating quantization bit quantities for combination coefficients based on their specific characteristics. Different component vectors (e.g., wideband vs. narrowband, different polarization directions) are allocated different bit quantities according to their impact on precoding performance, rather than using a uniform quantization scheme for all coefficients.
Solution Approach 2:
The patent changes the parameter of quantization bit quantity from a uniform value to a differentiated value based on component vector characteristics. The receive-end device determines appropriate bit quantities for different combination coefficients and feeds back this configuration to the transmit-end device, which then uses these differentiated parameters for quantization.
2Measurement precision
If more quantization bits are allocated to all coefficients, then the precoding vector precision improves, but the feedback overhead increases
Solution Approach 1:
Instead of uniformly increasing bit quantities for all coefficients, the patent applies local quality by selectively allocating more bits only to combination coefficients that have greater impact on precoding performance (such as wideband amplitude coefficients), while using fewer bits for less critical coefficients.
Solution Approach 2:
The patent applies partial action by not uniformly increasing precision for all coefficients, but rather applying enhanced precision only where necessary for critical component vectors, achieving sufficient overall performance without the overhead of universal high-precision quantization.
3Quantity of substance
If fewer quantization bits are used for combination coefficients, then the feedback overhead is reduced, but the precoding vector precision deteriorates
Solution Approach 1:
The patent resolves this contradiction by applying local quality - using fewer bits for less critical coefficients (such as narrowband phase coefficients) while maintaining higher bit quantities for critical coefficients (such as wideband amplitude coefficients), thereby reducing overall overhead while preserving essential precision.
Solution Approach 2:
The patent changes the quantization parameter allocation from a uniform high-precision scheme to a differentiated scheme where bit quantities are adjusted according to the importance and characteristics of each combination coefficient, achieving overhead reduction without sacrificing critical precision.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of this application disclose a method for indicating and determining a precoding vector, and a device, relate to precoding technologies, and help achieve the following beneficial effects: Precision of a precoding vector is ensured, and overheads for feeding back indication information indicating a combination coefficient are properly set, thereby increasing an overall system performance gain. The method may include: generating indication information, and sending the indication information. The indication information is used to indicate a plurality of component vectors of an ideal precoding vector and a combination coefficient of each component vector, the combination coefficient of the component vector includes at least one of the following types: a wideband amplitude coefficient, a narrowband amplitude coefficient, and a narrowband phase coefficient, and quantization bit quantities of same-type combination coefficients of at least two component vectors in the plurality of component vectors except a normalization reference component vector are different.