LDPC Base Matrix Extension for Flexible Code Rates and Row Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing LDPC codes, such as those used in 802.11ay and NR, lack flexibility in code rate and do not support parallel decoding of whole rows, limiting their performance in communication systems.
Innovation Solution
An LDPC code-based communication method that utilizes a base matrix derived from a storage matrix through indication information, allowing for flexible low-code-rate extensions by maintaining optimal edge density and reducing hardware complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed check matrix is stored for each code rate, then decoding can be performed, but flexible code rate adaptation is not supported
Solution Approach 1:
The patent uses a single base matrix that can generate multiple check matrices through row operations and column permutations. This universal base matrix serves multiple code rates, eliminating the need to store separate check matrices for each code rate while maintaining flexible code rate adaptation.
Solution Approach 2:
The patent changes parameters of the base matrix through row operations (adding rows, eliminating rows) and column permutations to generate different check matrices for different code rates. By modifying the matrix parameters dynamically, the system achieves code rate flexibility without storing multiple fixed matrices.
2Productivity
If the core matrix supports only parallel decoding of QC blocks, then hardware complexity is reduced, but parallel decoding of whole rows is not supported
Solution Approach 1:
The patent segments the decoding process into two levels: QC block-level parallel decoding for hardware efficiency, and row-level parallel decoding through check matrix construction. The check matrix is designed with row structures that enable simultaneous processing of multiple rows while maintaining QC block segmentation for practical hardware implementation.
3Adaptability or versatility
If low-code-rate extension is performed using conventional methods, then code rate can be reduced, but edge density optimization is not achieved
Solution Approach 1:
The patent applies local quality optimization by selectively adding rows to the check matrix based on edge density requirements. Different rows are added with specific patterns to maintain optimal edge density in different parts of the matrix, ensuring both low code rate support and reliable decoding performance.
Solution Approach 2:
The patent dynamically adjusts the check matrix structure during low-code-rate extension by selectively adding rows and performing column permutations. This dynamic construction process optimizes edge density adaptively rather than using fixed conventional extension methods, achieving both code rate flexibility and decoding reliability.
Data Source
AI summary
This application provides a method for constructing an LDPC base matrix. The LDPC base matrix may be obtained based on a storage matrix and indication information. Specifically, in a process of obtaining the LDPC base matrix, low-code-rate extension includes two manners: conventional extension and split extension. Whether each extension is conventional extension or split extension may be determined by the indication information.