Dual-Decoder LDPC Architecture for NAND Flash Latency Reduction
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
The implementation of 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 for early termination of LDPC iterative decoding based on syndrome weight and iteration thresholds, reducing the number of decoding iterations and improving overall performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If iterative decoding is performed to correct errors effectively, then error correction capability is improved, but decoding latency and processing time increase
Solution Approach 1:
The patent implements dynamic iteration termination by monitoring syndrome weight during the iterative decoding process. The decoder adaptively stops iterations when the syndrome weight indicates sufficient error correction, rather than executing a fixed number of iterations. This dynamic approach allows the system to maintain high error correction capability while significantly reducing latency when errors are few or already corrected, thereby resolving the contradiction between reliability and time loss.
Solution Approach 2:
The patent changes the decoding parameter (number of iterations) based on the syndrome weight threshold. When syndrome weight falls below the threshold, iterations are terminated early; otherwise, full iterations continue. This parameter adjustment strategy maintains error correction effectiveness while optimizing decoding speed, effectively balancing reliability and latency requirements.
2Reliability
If a large number of decoding iterations are performed, then error correction effectiveness is improved, but processing burden and computational complexity increase
Solution Approach 1:
The patent applies partial action by performing only the necessary number of decoding iterations required to achieve error correction, rather than always executing the maximum number of iterations. The syndrome weight threshold mechanism allows the system to stop iterations early when error correction is sufficient, reducing computational burden and processing complexity while maintaining effective error correction when needed.
Solution Approach 2:
The patent implements feedback control by continuously monitoring the syndrome weight during iterative decoding and using this information to determine whether to continue or terminate iterations. This feedback mechanism ensures that processing resources are allocated efficiently - full iterations are performed only when necessary, thereby reducing overall processing burden while maintaining error correction effectiveness.
3Loss of time
If early termination of iterative decoding is implemented, then decoding latency is reduced, but error correction capability may be compromised
Solution Approach 1:
The patent replaces the mechanical approach of fixed iterative decoding with a syndrome-weight-based termination mechanism. Instead of blindly executing a predetermined number of iterations, the system uses the syndrome weight (a mathematical indicator of error presence) to intelligently determine when to stop. This substitution ensures that early termination does not compromise error correction capability, as iterations continue until the syndrome weight confirms sufficient correction, thereby maintaining reliability while reducing latency.
Solution Approach 2:
The patent performs preliminary assessment of error correction status by calculating and monitoring syndrome weight at each iteration. This preliminary action allows the system to predict whether additional iterations are necessary before committing computational resources, enabling early termination when errors are already corrected while ensuring continued iterations when errors persist, thus balancing latency reduction with error correction capability.
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.


