Layered LDPC Decoding With Degree-Based Check Node Selection

Solution Overview

Problem

Existing LDPC decoding methods face challenges in achieving a balance between complexity and performance, with conventional sum-product algorithms requiring high computational complexity and simplified algorithms compromising decoding accuracy.

Innovation Solution

Applying a sum-product algorithm and an approximated sum-product algorithm, such as the min-sum algorithm, selectively to different check nodes based on their degree of connectivity, reducing complexity while maintaining performance by prioritizing CNs with lower degrees and considering shortened or punctured VNs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the sum-product algorithm is applied to all check nodes, then decoding accuracy is improved, but computational complexity increases

Engineering Contradiction:

Improvedecoding accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies different decoding algorithms to different check nodes based on their local properties (degree values). Specifically, check nodes with degree greater than a threshold use the sum-product algorithm for high accuracy, while check nodes with degree less than or equal to the threshold use the min-sum algorithm for lower complexity. This local differentiation resolves the contradiction by optimizing accuracy where needed and reducing complexity where possible.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The set of check nodes is segmented into two groups based on their degree values relative to a threshold. This segmentation allows the system to apply different algorithms to different segments, achieving a balance between overall decoding accuracy and computational complexity by not uniformly applying the high-complexity sum-product algorithm to all nodes.

Inventive Principle:

Principle #1Segmentation

2Device complexity

If the min-sum algorithm is applied to all check nodes, then computational complexity is reduced, but decoding accuracy deteriorates

Engineering Contradiction:

Improvecomputational complexityVSAvoiddecoding accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies different decoding algorithms to different check nodes based on their local properties (degree values). Specifically, check nodes with degree greater than a threshold use the sum-product algorithm for high accuracy, while check nodes with degree less than or equal to the threshold use the min-sum algorithm for lower complexity. This local differentiation resolves the contradiction by optimizing accuracy where needed and reducing complexity where possible.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The set of check nodes is segmented into two groups based on their degree values relative to a threshold. This segmentation allows the system to apply different algorithms to different segments, achieving a balance between overall decoding accuracy and computational complexity by not uniformly applying the high-complexity sum-product algorithm to all nodes.

Inventive Principle:

Principle #1Segmentation

3Productivity

If layered decoding with selective algorithm application is used, then the trade-off between complexity and performance is improved, but the decoding process becomes more complex to implement

Engineering Contradiction:

Improvedecoding efficiencyVSAvoidimplementation complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary classification of check nodes based on their degree values before the actual decoding process. By pre-calculating which check nodes should use sum-product algorithm and which should use min-sum algorithm, the system simplifies the implementation complexity during runtime while maintaining improved decoding efficiency through selective algorithm application.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20260081621A1Layered decoding of low density parity check (LDPC) codes

Publication Date: 2026.03.19 SAMSUNG ELECTRONICS CO LTD

AI summary

Layered decoding of low-density parity check (LDPC) codes are decoded by selecting one or more of a plurality of check nodes (CNs) based on degree in priority, where the degree in priority for each of the plurality of CNs is based on a number of variable nodes (VNs) connected to the respective CN. Layered decoding for an LDPC code is performed based on applying a sum-product algorithm and an approximated sum-product algorithm. The sum-product algorithm is applied to at least some of the selected one or more CNs among the plurality of CNs. The approximated sum-product algorithm is applied to one or more remaining CNs, other than the selected one or more CNs, among the plurality of CNs.