QC-LDPC Base Matrix Layout for Parallel Decoding Through Puncturing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing channel coding techniques, despite their good performance, face challenges in achieving high data throughput while efficiently managing encoding and decoding resources.
Innovation Solution
The method involves creating a base matrix for an irregular QC-LDPC code, where entries represent blocks of the matrix, and specific columns are selected to form groups of orthogonal rows, with puncturing of high-weight columns to enable layered and flooding decoding, achieving high parallelism during decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If puncturing of high-weight columns is applied to enable layered decoding, then parallelism during decoding is improved, but code complexity increases
Solution Approach 1:
The parity check matrix is segmented into high-weight columns and low-weight columns, with corresponding row divisions into first and second sets. This segmentation enables different decoding strategies for different parts of the matrix, achieving parallelism through structured division while managing complexity through organized resource allocation.
Solution Approach 2:
The decoding process dynamically adapts by applying layered decoding to groups of orthogonal rows from the second set and flooding decoding to the first set. This dynamic approach allows the system to optimize parallelism based on the structural properties of the punctured matrix, improving throughput while controlling resource usage through adaptive processing.
2Productivity
If columns are selected to form groups of orthogonal rows, then parallelism during decoding is improved, but the number of punctured information bits increases
Solution Approach 1:
Different columns are treated with different qualities - high-weight columns are punctured to enable orthogonal row grouping and parallelism, while low-weight columns are preserved to minimize information loss. This local differentiation optimizes the trade-off between parallelism and information retention by applying different strategies to different parts of the matrix.
Solution Approach 2:
The weight parameter of columns is used as a criterion for selection - columns with weight above a threshold are punctured, while those below are preserved. This parameter-based approach systematically balances parallelism improvement against information loss by changing the structural parameters of the code based on column weights.
3Productivity
If rows are divided into high-density and low-density sets, then decoding efficiency is improved, but matrix structure complexity increases
Solution Approach 1:
The rows of the parity check matrix are segmented into a first set (high-density) and a second set (low-density) based on weight thresholds. This segmentation simplifies decoding by allowing specialized processing for each set - flooding decoding for high-density rows and layered decoding for low-density rows - thereby improving efficiency while managing structural complexity through organized classification.
Solution Approach 2:
Different decoding qualities are applied to different row sets - the first set receives flooding decoding while the second set receives layered decoding. This local quality approach optimizes decoding efficiency by matching the decoding strategy to the density characteristics of each row set, improving overall performance while maintaining manageable matrix structure through differentiated processing.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Provided is an efficiently decodable QC-LDPC code which is based on a base matrix of an irregular QC-LDPC matrix, the base matrix being formed by columns and rows, the columns being dividable into one or more columns corresponding to punctured variable nodes (i.e. variable nodes corresponding to information bits which are used by the encoder but are not transmitted to or effectively treated as not received by the decoder) and columns corresponding to not-punctured variable nodes, and the rows being dividable into high-density rows (i.e. rows having a weight which is above a first weight) and low-density rows (i.e. rows having a weight which is below a second weight, wherein the second weight is equal to or smaller than the first weight), wherein a matrix defined by the overlap of the low-density rows and the columns corresponding to the not-punctured variable nodes is dividable into groups of orthogonal rows.