Non-Volatile Memory Soft-Bit Decoding for Read Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile memory devices often encounter erroneous or corrupted data due to noise and charge neutrality issues, leading to unreliable read-back states, which existing error correction methods struggle to address effectively.
Innovation Solution
The implementation of iterative probabilistic decoding using low density parity check codes and logarithmic likelihood ratios, combined with soft bit reading techniques that adjust reliability metrics and provide additional information through multiple subsets of compare levels to improve decoding convergence and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional error correction methods are used, then the system is simple to implement, but the reliability of data read-back is insufficient due to noise and charge neutrality issues
Solution Approach 1:
The patent transforms hard bit read data into soft bit data by changing the parameter representation from binary (0/1) to continuous probability values (log-likelihood ratios). This parameter transformation enables iterative probabilistic decoding to effectively correct errors caused by noise and charge neutrality issues, significantly improving data read-back reliability while managing system complexity through algorithmic rather than hardware-based solutions.
Solution Approach 2:
The patent implements iterative probabilistic decoding where soft bit data is repeatedly processed through decoding algorithms, with each iteration using feedback from previous decoding attempts to refine error correction. This feedback mechanism allows the system to progressively improve decoding accuracy by re-evaluating uncertain bits based on accumulated probability information from multiple read operations and decoding passes.
2Measurement precision
If soft bit reading techniques are implemented, then decoding accuracy is improved, but the number of read operations and processing time increase
Solution Approach 1:
The patent performs preliminary soft bit reading operations to gather probability information before final decoding. By pre-collecting soft bit data from multiple read operations and converting them to log-likelihood ratios in advance, the system prepares high-quality input for the iterative decoder, improving decoding accuracy while managing time through structured pre-processing rather than ad-hoc operations.
Solution Approach 2:
The patent employs periodic iterative decoding cycles where soft bit data is processed through multiple decoding passes at regular intervals. Each periodic iteration refines the probability estimates and corrects errors progressively, allowing the system to achieve high decoding accuracy through repeated cycles rather than single-pass processing, thereby balancing time consumption with improved precision.
Data Source
AI summary
Data stored in non-volatile storage is decoded using iterative probabilistic decoding. An error correcting code such as a low density parity check code may be used. In one approach, initial reliability metrics, such as logarithmic likelihood ratios, are used in decoding sensed states of a set of non-volatile storage element. The decoding attempts to converge by adjusting the reliability metrics for bits in code words which represent the sensed state. Soft data bits are read from the memory if the decoding fails to converge. Initial reliability metric values are provided after receiving the hard read results and at each phase of the soft bit operation(s). In one embodiment, a second soft bit is read from the memory using multiple subsets of soft bit compare levels. While reading at the second subset of compare levels, decoding can be performed based on the first subset data.