QC-LDPC Encoding Matrix Layout for Low Error Floors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computing methods for QC-LDPC codes do not optimize the check matrix effectively, leading to suboptimal error floors and inefficient use of storage space.
Innovation Solution
A novel encoding method using a first matrix with cyclic shift matrices and adjusted lifting factors to optimize the QC-LDPC code, reducing bit-error-rate floors and storage requirements while maintaining performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing computing methods are used for QC-LDPC check matrices, then implementation is straightforward, but bit-error-rate floors are suboptimal and storage space is inefficiently used
Solution Approach 1:
The patent modifies the check matrix by changing parameters including lifting factors and offset values of cyclic shift matrices. Specifically, it uses a base matrix with modified lifting factors (e.g., z1=16, z2=24, z3=32, z4=40) and adjusts offset values to optimize error floor performance while reducing storage requirements for the check matrix
Solution Approach 2:
The check matrix is divided into multiple block matrices of size z×z, where each block is either an all-zero matrix or a cyclic shift matrix. This segmentation allows independent optimization of each block's offset value to achieve better error floor performance with reduced storage
2Reliability
If lifting factors are increased to improve code performance, then error correction capability improves, but storage space requirements increase
Solution Approach 1:
The patent uses differentiated lifting factors for different block matrices (z1=16, z2=24, z3=32, z4=40) rather than uniform lifting factors, allowing optimization of error correction capability for different code rates and lengths while minimizing overall storage requirements through selective lifting factor assignment
Solution Approach 2:
The system dynamically selects lifting factors and offset values based on required code length and code rate. The check matrix structure adapts to different communication scenarios by adjusting which block matrices are active and their corresponding lifting factors, optimizing the balance between performance and storage
Data Source
Figure 1~2
Figure 3a
Figure 3b
AI summary
Embodiments of this application disclose an encoding method, a communication method, and an apparatus. A first information sequence is encoded by using a first matrix to obtain a second information sequence, where the first matrix meets a function related to a lifting factor and an element in Emax(H). In this way, a relatively low bit-error-rate floor can be maintained, and storage space occupied can be reduced.