Adaptive Kronecker MIMO Precoding for 2D Antenna Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional MIMO precoding schemes face challenges in reducing computational complexity and CSI overhead, especially in scenarios with degraded CSI quality due to user mobility or insufficient pilot transmission, and struggle to achieve optimal tradeoff between diversity and throughput in multi-user and single-user scenarios.
Innovation Solution
The implementation of an adaptive Kronecker product MIMO precoding scheme that allows for the sharing of precoding coefficients across sub-arrays of antenna elements, enabling efficient control of the tradeoff between diversity and throughput by adjusting the Kronecker product tradeoff parameter L, which reduces computational complexity and CSI feedback overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional MIMO precoding schemes are used with 2D antenna arrays, then beamforming capability is achieved, but computational complexity and CSI overhead increase significantly
Solution Approach 1:
The patent segments the 2D antenna array into multiple sub-arrays along one dimension (e.g., dividing columns into groups). This segmentation enables the precoding matrix to be constructed as a Kronecker product of two smaller matrices, reducing computational complexity from O(Nh*Nv) to O((Nh/L)*Nv + Nh*(Nv/L)) where L is the number of sub-arrays. The segmentation principle directly addresses the complexity issue while maintaining beamforming capability through the structured precoding design.
2Reliability
If conventional MIMO precoding schemes are used with 2D antenna arrays, then beamforming capability is achieved, but CSI feedback overhead increases
Solution Approach 1:
By segmenting the antenna array into sub-arrays and using Kronecker product structure, the patent reduces the number of independent CSI parameters that need to be feedback. Instead of feedbacking a full Nh*Nv precoding matrix, the system only needs to feedback parameters for the smaller constituent matrices, significantly reducing CSI feedback overhead while maintaining the beamforming capability through the Kronecker product structure.
3Device complexity
If Kronecker product MIMO precoding is used, then computational complexity is reduced, but performance degrades when CSI quality is poor due to user mobility
Solution Approach 1:
The patent introduces dynamic adaptation mechanisms where the system can switch between different precoding modes (Kronecker product-based and conventional) based on CSI quality assessment. When CSI quality is good, the low-complexity Kronecker product precoding is used; when CSI quality degrades due to user mobility, the system dynamically switches to more robust conventional precoding or adjusts the Kronecker product parameters to maintain performance. This dynamic adaptation resolves the contradiction between complexity reduction and performance reliability.
4Loss of information
If Kronecker product MIMO precoding is used, then CSI overhead is reduced, but tradeoff between diversity and throughput cannot be optimized
Solution Approach 1:
The patent introduces a tradeoff parameter L that controls the degree of Kronecker product structure used in precoding. By adjusting L (the number of sub-arrays), the system can optimize the balance between diversity and throughput for different operational conditions. This parameter change mechanism allows fine-grained control over the tradeoff while maintaining reduced CSI overhead, resolving the contradiction between overhead reduction and adaptability.
Data Source
AI summary
A network entity comprises a plurality of antenna elements arranged in one or more two dimensional (2D) arrays having one or more columns and rows. The network entity configured to determine at least one set of one or more precoding vectors related to the plurality of antenna elements, wherein each set of precoding vectors is associated with a different Kronecker product tradeoff parameter L≥1; and transmit, at least one set of a plurality of Cell Specific Reference Signals (CRS) to be used to estimate channel state information (CSI) based on the at least one set of precoding vectors and/or at least one Kronecker product tradeoff parameter L.