NAND Flash LDPC Decoding with Dual-Decoder Early Termination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current error correction systems in NAND flash storage, particularly LDPC codes, face challenges in latency and processing performance due to iterative decoding processes that may not meet the requirements of latency-sensitive or processing-sensitive applications, as they often require a large number of iterations to correct errors effectively.
Innovation Solution
A data processing system with a dual-decoder architecture that includes a first decoder for normal decoding and a second decoder for fast decoding, allowing early termination of LDPC iterative decoding based on syndrome weight and iteration thresholds, reducing the number of iterations needed for error correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If iterative decoding is performed to ensure error correction performance, then reliability is improved, but loss of time increases due to multiple iterations required
Solution Approach 1:
The patent implements dynamic decoding by allowing the decoding process to adaptively terminate based on real-time syndrome weight evaluation. The system transitions from fixed iterative decoding to dynamic decoding where the number of iterations is adjusted according to the actual error condition, enabling early termination when errors are few and simple patterns are detected, thus reducing latency while maintaining reliability
Solution Approach 2:
The patent changes the decoding parameter (number of iterations) based on syndrome weight thresholds. By monitoring syndrome weight during iteration and comparing against predefined thresholds, the system dynamically adjusts the iteration count parameter, allowing fewer iterations for low-weight syndromes and more iterations for high-weight syndromes, thereby optimizing the balance between reliability and time loss
2Reliability
If full iterative decoding is performed for all data chunks, then error correction performance is maintained, but productivity decreases due to processing burden
Solution Approach 1:
The patent applies partial action by performing full iterative decoding only when necessary (high syndrome weight) and using simplified early termination when sufficient (low syndrome weight). This partial application of full decoding based on condition evaluation maintains error correction capability for difficult cases while improving overall processing throughput by avoiding unnecessary full iterations for easier cases
Solution Approach 2:
The patent implements feedback mechanisms where syndrome weight results from each iteration are fed back to determine whether to continue or terminate decoding. This feedback loop allows the system to evaluate decoding progress in real-time and make informed decisions about continuing full iterative processing or early termination, thereby optimizing productivity while maintaining reliability
3Reliability
If normal decoding is performed for all subsequent data chunks after a failure, then error correction is ensured, but loss of time increases due to redundant processing
Solution Approach 1:
The patent applies preliminary action by evaluating syndrome weight thresholds before committing to full iterative decoding for subsequent data chunks. By pre-establishing thresholds and evaluating syndrome weight in advance, the system can determine early whether full decoding is necessary or if early termination is sufficient, avoiding redundant processing time while ensuring error correction when truly needed
Solution Approach 2:
The patent segments the decoding process into distinct phases: initial syndrome weight evaluation, threshold comparison, and conditional decoding execution. This segmentation allows the system to handle different data chunks with appropriate decoding intensity based on their specific error characteristics, reducing redundant processing time while maintaining reliability through targeted full decoding when required
Data Source
AI summary
A data processing system includes a storage medium, and a controller including a data processing block, configured to receive data from a host, transmit the received data to the storage medium, read data from the storage medium in response to a read request from the host, and decode the read data by the data processing block according to multiple decoding modes. The data processing block includes a first decoder and a second decoder, and is configured to manage the first decoder and the second decoder to run the decoding for the read data, and activate a fast decoding having shorter latency than a normal decoding after a fast decoding condition is satisfied.