LDPC Base Matrix Selection for Error Floor and Decoding Throughput
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current LDPC codes face challenges such as error floor and limited decoding throughput, which hinder their use in ultra-high reliability transmission and limit the maximum parallelism in hierarchical decoding.
Innovation Solution
The proposed method involves using a base matrix for LDPC coding and decoding that includes elements indicating an all-zero square matrix and a cyclic shift of an identity matrix, with subsets of base matrices having the same coding rate, to improve error floor performance and decoding speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional LDPC codes are used, then decoding complexity is reduced, but error floor performance deteriorates and reliability is limited
Solution Approach 1:
The patent applies parameter changes by modifying the base matrix structure to include cyclic shift elements and organizing base matrices into subsets with different coding rates. This transforms the conventional LDPC code parameters to achieve better error floor performance while maintaining decoding complexity through the sparse structure.
Solution Approach 2:
The patent segments the base matrix into multiple subsets where each subset contains base matrices with the same coding rate. This segmentation allows the system to select appropriate subsets for different transmission conditions, improving reliability without significantly increasing overall decoding complexity.
2Productivity
If maximum parallelism is adopted in hierarchical decoding, then decoding throughput is increased, but error floor performance deteriorates
Solution Approach 1:
The patent introduces dynamics by allowing the selection of different base matrix subsets based on transmission conditions. The system can dynamically adjust the degree of parallelism in hierarchical decoding by choosing appropriate subsets, thereby achieving both high throughput and good error floor performance depending on channel conditions.
Solution Approach 2:
The patent changes the structural parameters of the base matrix to include cyclic shift elements and organizes them into subsets with different coding rates. This enables the system to optimize the balance between parallelism and error floor performance by selecting appropriate parameter configurations for different operating conditions.
3Reliability
If base matrix is optimized for ultra-high reliability, then error floor performance is improved, but device complexity increases
Solution Approach 1:
The patent segments the base matrix into multiple subsets with different coding rates and organizes them systematically. This segmentation allows the system to achieve ultra-high reliability by selecting appropriate subsets without implementing a single overly complex base matrix, thereby managing device complexity through modular organization.
Solution Approach 2:
The patent creates a universal base matrix structure that can serve multiple functions by including cyclic shift elements and organizing matrices into subsets. This universal structure can adapt to different reliability requirements and transmission conditions, achieving ultra-high reliability without proportionally increasing device complexity through its multi-functional design.
Data Source
AI summary
The present disclosure provides coding and decoding methods, a communication device, and a storage medium. The coding method, applied to a first transmission node, includes: determining a base matrix for low density parity check (LDPC) coding from a preset base matrix set; performing LDPC coding on original data according to the base matrix, so as to obtain coded data; transmitting the coded data to a second transmission node; wherein the base matrix includes two kinds of elements: an element indicating an all-zero square matrix, and an element indicating a cyclic shift of an identity matrix; the base matrix set includes at least two non-empty subsets of base matrices, and all base matrices in each subset of the at least two non-empty subsets of base matrices have a same coding rate.