NAND Flash Read Voltage Selection Using Abnormal Gray Code Counts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Rewritable non-volatile memory modules, such as NAND flash memory, face challenges with unstable threshold voltages, leading to inaccurate measurement of threshold voltage distributions and reduced reading efficiency due to unstable physical states of memory cells.
Innovation Solution
A data reading method that uses multiple read voltage sets to determine optimized read voltages by sorting and updating Gray code indexes, identifying abnormal Gray code counts, and selecting optimized read voltages to improve read voltage accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple read voltage sets are used to measure threshold voltage distributions, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the read voltage measurement process into multiple discrete voltage sets (X read voltage sets with N voltages each), where each set targets specific threshold voltage ranges. This segmentation allows systematic measurement of different portions of the threshold voltage distribution, improving overall measurement precision while maintaining organized process control.
Solution Approach 2:
The patent performs preliminary actions by first measuring threshold voltage distributions using multiple read voltage sets before final optimization. The abnormal Gray code count sets are calculated in advance, and the most abnormal count is identified beforehand, allowing the system to pre-determine which read voltages need optimization rather than doing so reactively.
2Measurement precision
If Gray code index updating and abnormal count calculation are performed for all memory cells, then measurement precision is improved, but productivity decreases
Solution Approach 1:
The patent extracts only the abnormal Gray code counts from the complete set of read results, focusing computational resources on identifying and analyzing deviations from expected patterns. By taking out only the abnormal cases for detailed analysis rather than processing all data uniformly, the system improves measurement precision while reducing overall processing burden.
Solution Approach 2:
The patent applies local quality by performing detailed Gray code index updating and abnormal count calculation only where needed - specifically for identifying threshold voltage distribution anomalies. The optimization is applied locally to specific read voltages that show abnormal patterns rather than uniformly across all voltages, improving precision where required while maintaining productivity.
3Measurement precision
If multiple read voltage sets with voltage offsets are used, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent employs periodic action by using multiple read voltage sets applied in sequence with regular voltage offsets. Each read voltage set is applied periodically to measure different portions of the threshold voltage distribution, allowing systematic data collection that improves precision while maintaining a structured, efficient measurement rhythm.
Solution Approach 2:
The patent performs preliminary measurement actions using the multiple read voltage sets with voltage offsets before final optimization. By conducting these measurements in advance and calculating abnormal Gray code count sets beforehand, the system prepares optimization data proactively, reducing the time required for actual read operations later.
Data Source
AI summary
A data reading method is provided. The method includes using X read voltage sets to read a target word line, so as to obtain X read results; in a first order, updating a final Gray code index of each of a plurality of target memory cells of the target word line, and obtaining (X−1) abnormal Gray code count sets according to the X read results, wherein an ith read result among the X read results includes a Gray code corresponding to an ith read voltage set of each of the target memory cells, and the Gray code corresponds to one of N Gray code indexes; and selecting (N−1) optimized read voltages from (X−1)*(N−1) read voltages of the corresponding (X−1) read voltage sets to form an optimized read voltage set according to the obtained (X−1) abnormal Gray code count sets.