ECC Decoder Control Using Soft Syndrome Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The continuous operation of nonvolatile memory degrades cell distribution, leading to an increased probability of obtaining data chunks with a large number of errors during read operations, even when using pre-set read voltages.
Innovation Solution
A controller is designed to improve ECC decoding success probability by generating hard and delta syndromes, and using these to determine whether to restore a soft data chunk, ultimately selecting the data chunk with the lowest syndrome weight for ECC decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a pre-set read voltage is used for continuous read operations, then the read operation speed is maintained, but the probability of obtaining data chunks with a large number of errors increases due to cell distribution degradation
Solution Approach 1:
The system performs preliminary syndrome calculations (hard syndrome and delta syndrome) and generates soft syndromes before the actual ECC decoding operation. By pre-calculating syndromes and determining syndrome weights, the system identifies data chunks with lower error rates in advance, selecting optimal candidates for decoding. This preliminary action increases the probability of successful ECC decoding while maintaining efficient read operation speed.
2Reliability
If traditional ECC decoding is performed on all data chunks, then decoding coverage is complete, but the success probability of ECC decoding decreases when cell distribution is degraded
Solution Approach 1:
The system segments the data chunk processing by dividing it into multiple stages: hard syndrome calculation, delta syndrome calculation, soft syndrome generation, syndrome weight determination, and selective decoding. This segmentation allows the system to process syndromes and identify optimal data chunks before performing the complex ECC decoding operation, thereby increasing decoding success probability while managing complexity through structured multi-stage processing.
Solution Approach 2:
The soft syndrome acts as an intermediary between the hard/delta syndromes and the final ECC decoding operation. By generating soft syndromes through logic operations on hard and delta syndromes, and by using syndrome weights as a selection criterion, the system introduces an intermediate evaluation layer that identifies promising data chunks before committing to full decoding, thus improving success probability without proportionally increasing overall complexity.
3Reliability
If multiple read operations are performed to find low-error data chunks, then the probability of successful ECC decoding increases, but the access operation time increases
Solution Approach 1:
The system performs preliminary syndrome calculations and soft syndrome generation in parallel or during idle periods before the actual read operation completes. By pre-calculating hard syndromes, delta syndromes, and soft syndromes, and by determining syndrome weights in advance, the system identifies optimal data chunks without requiring additional sequential read operations, thereby increasing decoding success probability while minimizing access time penalty.
Data Source
AI summary
The present technology provides a controller for controlling a memory device comprising: a hard syndrome calculator configured to generate a hard syndrome of a hard data chunk received from the memory device; a delta syndrome calculator configured to generate a delta syndrome of a delta bit data received from the memory device, the delta bit data indicating a reliability of the hard data chunk; a soft syndrome generator configured to generate a soft syndrome of the hard syndrome and the delta syndrome; a data determinator configured to select, as hard decision data, one of the hard data chunk and a soft data chunk, the selected data chunk corresponding to a syndrome having a lowest syndrome weight; and an error corrector configured to perform an ECC decoding operation on the hard decision data.