Cross-Directional ECC Decoding for Nonvolatile Data Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage devices using nonvolatile memory struggle with improving error correction capabilities, as existing error correction methods are not effectively interdependent, leading to incomplete data recovery from errors in data groups.
Innovation Solution
The proposed method involves performing interdependent error correction encoding and decoding operations for data groups in both directions, using first and second encoding operations to generate parity data and correct errors in data groups, ensuring that if one decoding operation fails, the other can succeed by utilizing shared data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single error correction decoding operation is performed for a data group, then the decoding process is simple and fast, but the error correction capability is insufficient when errors exceed the correction threshold
Solution Approach 1:
The data group is divided into multiple sub-data-groups, and the error correction decoding is performed separately for each sub-data-group. This segmentation allows the system to correct errors in a distributed manner, improving overall error correction capability while keeping each individual decoding operation manageable in complexity.
Solution Approach 2:
The patent introduces a multi-dimensional decoding approach by performing decoding operations in different directions or layers. When the first decoding operation fails, a second decoding operation is executed from another dimension, providing an additional pathway for error correction without requiring a complete redesign of the decoding architecture.
2Reliability
If multiple error correction decoding operations are performed for data groups, then the error correction capability is improved, but the processing time and computational resources increase
Solution Approach 1:
The system performs a first error correction decoding operation as a preliminary step before attempting more complex decoding methods. This preliminary action quickly handles cases with fewer errors, and only when this initial decoding fails does the system proceed to the second decoding operation, thereby reducing average processing time while maintaining high error correction capability.
Solution Approach 2:
The patent implements a hierarchical decoding strategy where the first decoding operation handles partial error correction cases efficiently. Only when this partial action is insufficient does the system activate the second decoding operation, applying computational resources proportionally to the actual error severity rather than always using the full decoding capability.
Data Source
AI summary
An error correction code processing method includes performing a first encoding operation for a data group of a first direction; performing a second encoding operation for a data group of a second direction, wherein the data group of the first direction shares one or more data with the data group of the second direction; performing a first decoding operation of correcting an error included in the data group of the first direction; and performing a second decoding operation of correcting an error included in the data group of the second direction when the first decoding operation fails.