Flash Memory Read Reference Voltage Estimation via Disparity Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing flash memory systems face challenges in accurately detecting stored data values due to physical non-uniformity across cells and changes in the flash read channel over time, leading to errors in bit error rate (BER), which are not effectively addressed by current adaptive channel tracking algorithms that rely on assumptions about Gaussian distributions.
Innovation Solution
An adaptive channel tracking algorithm that calculates disparity and derivative metrics by repeatedly reading flash memory pages with multiple reference voltages, generating a disparity vector and its derivative, to estimate an optimal read reference voltage without relying on assumptions about underlying cell voltage distributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If adaptive channel tracking algorithms assume Gaussian or Gaussian-like write voltage distributions, then the estimation of optimal read reference voltage can be performed, but the algorithms become relatively complex and computationally intensive
Solution Approach 1:
The patent extracts the essential information needed for read reference voltage estimation by taking out only the necessary statistical moments (mean and variance) from the cell voltage distributions, rather than assuming complete Gaussian distributions. This extraction approach simplifies the algorithm by focusing only on the critical parameters needed for accurate estimation, eliminating the computational burden of full Gaussian assumption-based algorithms.
Solution Approach 2:
The patent changes the parameter representation from assuming full Gaussian distribution parameters to using only the first two moments (mean and variance) of the cell voltage distributions. This parameter change simplifies the mathematical model while maintaining estimation accuracy, as the patent derives read reference voltage estimates based on moment matching rather than full distributional assumptions.
2Measurement precision
If adaptive channel tracking algorithms rely on assumptions about Gaussian distributions, then read reference voltage can be estimated, but performance penalties occur when the assumptions are not accurate
Solution Approach 1:
The patent segments the distribution characterization into discrete moment calculations rather than relying on a single Gaussian assumption. By calculating the first and second moments separately and using these to estimate read reference voltages, the method adapts to the actual distribution shape without being constrained by Gaussian assumptions, improving reliability across different operating conditions.
Solution Approach 2:
The patent changes from fixed distributional assumptions to dynamic moment-based parameters that are calculated from actual read data. This allows the estimation method to adapt to changing cell voltage distributions caused by various degradation mechanisms, maintaining performance consistency without relying on inaccurate Gaussian assumptions.
3Measurement precision
If multiple read reference voltages are provided for MLC or TLC flash memory, then data detection accuracy improves, but the system complexity and computational requirements increase
Solution Approach 1:
The patent segments the problem of determining multiple read reference voltages by treating each voltage level independently through moment calculations. By calculating the first and second moments for each cell state and deriving read reference voltages from these segmented moment values, the system manages complexity while maintaining accurate detection for MLC and TLC memories.
Solution Approach 2:
The patent changes the approach from providing multiple fixed reference voltages to dynamically estimating read reference voltages based on calculated moment parameters. This parameter change allows the system to adapt the reference voltages to actual cell conditions, improving detection accuracy while reducing the complexity of manual voltage calibration and management.
Data Source
AI summary
An adaptive channel tracking algorithm performed by a flash memory system obtains disparity metrics and derivative metrics and uses a combination of the disparity and derivative metrics to estimate an optimal read reference voltage. The estimation of the optimal read reference voltage does not rely on assumptions about the underlying cell voltage distributions and results in a good estimate of the read reference voltage even if the standard deviations of the cell voltage distributions are different. In addition, the algorithm is relatively simple and less computationally intensive to perform than the known tracking algorithms.