Flash Memory Read Voltage Determination via Histogram Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining a reliable sense parameter, such as a read voltage level, in multilevel cell flash memories is challenging due to noise, fluctuations, and distribution widths, which affects the accuracy of data states and requires improved methods for filtering and adjusting parameters.
Innovation Solution
A method involving creating a histogram of sensed data, filtering it using averaging or weighted averaging, determining local minimums, and adjusting the read voltage level based on the average of the lowest and highest local minimums, with optional sigma-weighted mean interpolation and offset adjustments to account for varying distribution widths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional histogram minimum detection is used to determine read voltage levels, then the process is simple, but noise and fluctuations cause inaccurate sense parameter determination
Solution Approach 1:
The patent applies preliminary action by performing histogram filtering and local minimum identification before final read voltage level determination. The method pre-processes the histogram data using smoothing algorithms and identifies candidate minimum points in advance, then selects the optimal read voltage level from these pre-identified candidates, improving accuracy without excessive complexity
Solution Approach 2:
The patent introduces an intermediary element - the filtered histogram with identified local minimums - between the raw sensed data and the final read voltage level determination. This intermediary processing step acts as a mediator that reduces noise effects while maintaining a manageable determination process
2Reliability
If filtering and adjustment operations are applied to histogram data, then noise effects are reduced and accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies partial action by implementing selective filtering and adjustment operations only where needed in the histogram data. Rather than uniformly processing all data points with complex algorithms, the method applies filtering to reduce noise and performs adjustments only for identified local minimums, achieving improved reliability with moderate computational complexity
Solution Approach 2:
The histogram data itself provides information for its own optimization - the local minimums in the histogram naturally indicate candidate read voltage levels. The filtering and adjustment processes serve to enhance these naturally occurring features rather than imposing external complex structures, allowing the data to guide its own optimization
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Memory devices and methods for operating a memory include filtering a histogram of sensed data of the memory, and adjusting a parameter used to sense the memory using the filtered histogram. Filtering can be accomplished by averaging or summing, and may include weighting the sums or averages.