LTE Channel Feedback Matrix Inverse Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for determining channel feedback information in LTE systems require high calculation amounts due to the need to calculate the inverse of positive definite Hermite matrices, particularly in the process of joint estimation of Precoding Matrix Indicator (PMI) and Rank Indication (RI), which is computationally intensive.
Innovation Solution
The method involves obtaining an (N-1)th layer interference-plus-noise matrix and using it to calculate the Nth layer matrix, reducing the computational complexity by adopting a simplified process to calculate SINR, specifically through the formula Jv=[Jv-1+t5-t4-t4Ht3], which results in a square magnitude calculation amount instead of cubic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the inverse of positive definite Hermite matrices is calculated using Cholesky decomposition to determine SINR values, then measurement precision is improved, but calculation complexity increases to cubic magnitude
Solution Approach 1:
The patent transforms the matrix inversion problem into a recursive calculation using previously computed interference-plus-noise matrices. Instead of directly inverting the positive definite Hermite matrix at each step, the method updates the interference-plus-noise matrix incrementally from layer N-1 to layer N, changing the computational parameters from direct inversion to recursive update, thereby reducing complexity from cubic to square magnitude while preserving SINR calculation precision
Solution Approach 2:
The patent performs preliminary calculation of the interference-plus-noise matrix for the first layer before computing SINR values for subsequent layers. By pre-computing and storing the interference-plus-noise matrix results from previous layers, the method avoids redundant matrix inversion operations, reducing overall calculation complexity while maintaining measurement precision for all layers
2Productivity
If SINR values for all layers and all precoding matrices are calculated to determine PMI and RI, then productivity is improved, but calculation amount increases significantly
Solution Approach 1:
The patent segments the channel feedback determination process into distinct stages: first calculating interference-plus-noise matrices for each layer separately using recursive updates, then computing SINR values based on these pre-computed matrices. This segmentation allows the calculation to proceed layer-by-layer with reused intermediate results, improving productivity while reducing the total calculation amount compared to computing all SINR values simultaneously without intermediate storage
Solution Approach 2:
The patent performs preliminary computation of interference-plus-noise matrices before SINR calculation. By pre-computing these matrices and storing them for reuse across different precoding matrix evaluations, the method reduces redundant calculations when determining PMI and RI, thereby improving productivity while controlling the quantity of computational operations required
Data Source
AI summary
A method and a device for determining channel feedback information in a Long Term Evolution (LTE) system are provided. The method includes: obtaining an (N−1)th layer interference-plus-noise matrix in an LTE system, and obtaining an Nth layer interference-plus-noise matrix according to the (N−1)th layer interference-plus-noise matrix; and then, calculating a Signal to Interference plus Noise Ratio (SINR) corresponding to the Nth layer interference-plus-noise matrix according to the Nth layer interference-plus-noise matrix, and determining channel feedback information according to the SINR. It can be seen that, in the embodiments of the present invention, in a processing manner, a current layer interference-plus-noise matrix is calculated according to an obtained previous layer interference-plus-noise matrix through calculation in a calculation process of the channel feedback information, thus effectively reducing the calculation amount in the calculation process of the channel feedback information.