SC-LDPC Min-Sum Scaling for Lower-Complexity Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional error correction code (ECC) designs for non-volatile memory, such as NAND flash memory, face challenges with increased hardware complexity and diminishing returns in raw bit error rate (RBER) improvement as block length increases, especially for spatially-coupled low-density parity-check (SC-LDPC) codes, which are difficult to implement due to high hardware complexity and limited RBER gains.
Innovation Solution
The implementation of non-uniform iteration-dependent min-sum scaling factors in min-sum decoders for SC-LDPC codes, which vary across check nodes and decoding iterations, synchronized with wave-like decoding behavior, and optimized using density evolution analysis to enhance RBER performance and memory capacity while reducing hardware complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional ECC decoding solutions are used for SC-LDPC codes, then hardware complexity is reduced, but raw bit error rate performance is degraded
Solution Approach 1:
The patent applies dynamics by making the scaling factors variable across different decoding iterations rather than static. The scaling factors are adjusted dynamically based on the iteration number to optimize RBER performance while maintaining hardware efficiency. This dynamic adjustment allows the system to achieve belief-propagation-like performance without the full hardware complexity.
Solution Approach 2:
The patent implements local quality by applying different scaling factors to different check nodes based on their local properties (degree, position in the Tanner graph). This localized customization of decoding parameters optimizes performance for specific code structures without requiring universal complex hardware, thus resolving the contradiction between reliability and device complexity.
2Quantity of substance
If block length is increased for SC-LDPC codes, then memory capacity increases, but hardware complexity increases and RBER improvement diminishes
Solution Approach 1:
The patent changes key parameters of the min-sum decoding algorithm, specifically introducing iteration-dependent scaling factors that vary with block length. This parameter optimization allows the system to maintain efficient hardware complexity while achieving improved RBER performance across different block lengths, thus supporting increased memory capacity without proportional hardware complexity increases.
3Reliability
If uniform scaling factors are used in min-sum decoding, then hardware complexity is minimized, but RBER performance is limited
Solution Approach 1:
The patent transitions from static uniform scaling factors to dynamic iteration-dependent scaling factors. This dynamic approach allows the system to adapt scaling factors based on decoding progress and code properties, achieving superior RBER performance while maintaining hardware efficiency through systematic rather than exhaustive customization.
Solution Approach 2:
The patent applies different scaling factors to different check nodes based on their local characteristics such as degree and position. This localized differentiation improves RBER performance by optimizing decoding for specific code structures without requiring full belief-propagation hardware complexity, thus resolving the contradiction between performance and complexity.
Data Source
AI summary
Systems, apparatuses and methods may provide for technology to receive a codeword containing an SC-LDPC code and conduct a min-sum decode of the SC-LDPC code based on a plurality of scaling factors. In an embodiment, the scaling factors are non-uniform across check nodes and multiple iterations of the min-sum decode.