Non-Volatile Storage Decoding With Multiple Read Reliability Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile storage devices face challenges in reliably decoding data due to noise, particularly 1/f noise from electron trapping and de-trapping in trap sites, which affects the accuracy of read operations.
Innovation Solution
A method involving multiple read operations and iterative probabilistic decoding using reliability metrics, such as logarithmic likelihood ratios, to improve decoding accuracy by adjusting and refining probability metrics based on subsequent sense operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple read operations are performed to improve decoding reliability, then data decoding reliability is improved, but read operation time and complexity increase
Solution Approach 1:
The patent applies preliminary action by performing multiple read operations before the decoding process to collect sufficient reliability metrics. By gathering read data in advance from multiple operations, the system prepares comprehensive reliability information that enables accurate probabilistic decoding, thereby improving decoding reliability while managing the time investment before the actual decoding begins.
Solution Approach 2:
The patent implements partial or excessive action by performing more read operations than the minimum single read typically required. By conducting multiple read operations (excessive action), the system collects redundant reliability metrics that exceed what a single read would provide, enabling more robust probabilistic decoding at the cost of increased read operation time.
2Measurement precision
If iterative probabilistic decoding is used to improve decoding accuracy, then decoding accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies feedback by implementing an iterative probabilistic decoding process where reliability metrics from multiple read operations are continuously incorporated and refined. The decoding algorithm uses feedback from each iteration to adjust probability metrics and converge toward the most accurate decoded state, improving decoding accuracy through repeated refinement of the decoding decisions based on accumulated reliability information.
Solution Approach 2:
The patent implements dynamics by using a probabilistic decoding approach where reliability metrics are dynamically adjusted throughout the iterative process. Rather than using fixed threshold values, the system dynamically updates probability metrics based on incoming read data, allowing the decoding process to adapt and converge optimally according to the specific reliability characteristics observed in the multiple read operations.
3Object-affected harmful factors
If reliability metrics from multiple reads are used to mitigate noise effects, then noise resistance is improved, but processing overhead increases
Solution Approach 1:
The patent applies merging by combining reliability metrics from multiple read operations into a unified probabilistic decoding framework. Instead of processing each read operation separately, the system merges the reliability information from all reads into a cohesive set of probability metrics that collectively represent the noise-resistant interpretation of the stored data, thereby mitigating noise effects through aggregation of multiple observations.
Data Source
AI summary
Data stored in non-volatile storage is decoded using iterative probabilistic decoding and multiple read operations to achieve greater reliability. 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 read data 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. If convergence does not occur, e.g., within a set time period, the state of the non-volatile storage element is sensed again, current values of the reliability metrics in the decoder are adjusted, and the decoding again attempts to converge.