Adaptive Voltage Read Threshold Generation for NAND Flash Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In NAND flash memory, as more bits are stored per memory cell, the threshold voltage ranges become smaller, leading to increased error rates in determining the memory cell's value, and existing error-correction codes struggle to adaptively determine optimal read threshold voltages due to the complexity of varying factors affecting these voltages over time.
Innovation Solution
A computer system with a processor and memory configured to generate voltage read thresholds using probability distributions and an objective function, which selects and adjusts voltage read ranges to improve data reading accuracy, reducing raw bit error rates and enhancing storage device performance by employing neural networks and efficient polynomial approximations to minimize computational operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If more bits are stored per memory cell, then storage capacity is improved, but error rate in determining memory cell value increases
Solution Approach 1:
The patent changes the parameters used for reading memory cells by transitioning from fixed voltage thresholds to adaptive probability distributions. By modeling threshold voltages as probability distributions with varying parameters (mean, standard deviation) based on multiple factors (cell location, program history, temperature), the system can dynamically adjust read parameters to maintain reliability while preserving high storage capacity in multi-bit cells.
2Measurement precision
If offline memory characterization with look-up tables is used to determine optimal read thresholds, then read threshold accuracy is improved, but device complexity and implementation difficulty increase
Solution Approach 1:
The patent extracts only the essential parameters needed to define probability distributions (mean, standard deviation, and key factor indicators) rather than storing complete look-up tables for all possible conditions. This extraction approach maintains read threshold accuracy by preserving the most significant variation factors while dramatically reducing memory requirements and implementation complexity compared to comprehensive look-up tables.
Solution Approach 2:
The patent implements dynamic adaptation by allowing the probability distribution parameters to change based on current operating conditions and cell history. Instead of static look-up table values, the system dynamically calculates distribution parameters using stored factor information and current read conditions, enabling adaptive optimization without requiring exhaustive pre-computation for all possible states.
3Measurement precision
If adaptive determination of read threshold voltages is implemented to account for varying factors, then read threshold accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary characterization during manufacturing and initial operation to establish the relationship between cell factors and probability distribution parameters. This pre-computation stores only the essential factor-parameter mappings in compact form, avoiding the need for complex real-time calculations during actual read operations. The system pre-identifies which factors (cell location, program count, temperature ranges) most significantly affect threshold voltages and stores corresponding distribution parameters for these identified factors.
Data Source
AI summary
Techniques related to methods and systems for improving a performance related to reading data stored in memory cells. The method includes selecting a first voltage read range and a second voltage read range from multiple voltage read ranges that are associated with a number of bits storable in a memory cell. The method includes receiving, a first set of parameters that represent a first probability distribution of first candidate voltage read thresholds within the first voltage read range. The method includes receiving a second set of parameters that represent a second probability distribution of second candidate voltage read thresholds within the second voltage read range. The method includes generating, based on an input to an objective function, a voltage read threshold. The method includes reading data stored in the memory cell based on the voltage read threshold.