Structured LDPC Encoding and Decoding With Reduced Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low Density Parity Check (LDPC) codes are not widely deployed due to complexity in encoding and decoding processes, high storage requirements, and challenges in connecting processing nodes, as well as the lack of guaranteed convergence in soft decision decoding.
Innovation Solution
A method and apparatus for efficiently encoding and decoding structured LDPC codes using a structured parity check matrix, where information is organized in tabular form to simplify operations and reduce storage needs, and employing parallelizable decoding algorithms with soft decision decoding to ensure convergence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LDPC encoding is implemented using generator matrix, then encoding capability is achieved, but storage requirements increase significantly due to large non-sparse matrix
Solution Approach 1:
The patent extracts only the essential structural information from the full generator matrix by using parity check matrices with quasi-cyclic structure. Instead of storing the complete large non-sparse generator matrix, the system stores compact representations that define the matrix structure through smaller base matrices and shift patterns, significantly reducing storage while maintaining encoding capability
Solution Approach 2:
The patent applies local quality by using different structural properties in different parts of the code. The parity check matrix is divided into quasi-cyclic blocks with specific local structures that can be efficiently stored and processed. Each block has localized patterns that reduce overall storage requirements while maintaining global error correction capability
2Reliability
If LDPC codes use large blocks to be effective, then error correction performance improves, but storage and processing complexity increases
Solution Approach 1:
The patent segments the large LDPC code into smaller quasi-cyclic blocks that can be processed independently or in parallel. The parity check matrix is divided into sub-matrices with regular structures, allowing the large block to be handled through multiple smaller processing units. This segmentation maintains the error correction performance of large blocks while reducing the complexity of individual processing steps
Solution Approach 2:
The patent introduces dynamic elements in the form of shift registers and cyclic shifts that allow the same base matrix structure to generate different parity check matrices for different code rates and block lengths. This dynamic approach enables a single compact representation to serve multiple large block configurations, reducing storage requirements while maintaining effectiveness
3Productivity
If parallel decoder engines are used, then decoding speed increases, but connection network complexity between processing nodes increases
Solution Approach 1:
The patent merges multiple decoder engines into a unified structure where processing nodes share common memory spaces and communication interfaces. The quasi-cyclic structure of the code allows parallel processing units to operate on different blocks simultaneously while accessing shared parity check matrix definitions, reducing the need for complex point-to-point connections between nodes
4Measurement precision
If soft decision decoding is implemented, then decoding accuracy improves, but convergence cannot be guaranteed
Solution Approach 1:
The patent implements feedback mechanisms in the iterative decoding process where decoding results are fed back to adjust subsequent processing steps. The structured parity check matrix enables feedback loops that monitor convergence criteria and adjust decoding parameters dynamically, ensuring that soft decision decoding converges to a valid solution while maintaining high accuracy
Data Source
Figure 1
Figure 2A~2B
Figure 2C~2D
AI summary
An approach is provided for processing structure Low Density Parity Check (LDPC) codes. Memory storing edge information and a posteriori probability information associated with a structured parity check matrix used to generate Low Density Parity Check (LDPC) coded signal are accessed. The edge information represent relationship between bit nodes and check nodes, and are stored according to a predetermined scheme that permits concurrent retrieval of a set of the edge information.