Analog Memory Cell Read Threshold Estimation via Distribution Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices face challenges in accurately setting read thresholds for analog memory cells due to varying programming level distributions and potential overlap between adjacent distributions, leading to read errors and inefficiencies.
Innovation Solution
A method and system that estimate the median of each programming level distribution using intermediate read thresholds positioned within the distribution, calculating an optimal read threshold based on these medians to minimize read errors and maximize accurate data retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If read thresholds are set at the edges or between programming level distributions, then the reading process is simple, but read errors increase due to overlap between adjacent distributions
Solution Approach 1:
The system performs preliminary reading operations using intermediate read thresholds to estimate the median of each programming level distribution before final data retrieval. This preliminary action enables the system to calculate optimal read thresholds based on actual distribution characteristics, resolving the contradiction between simple threshold setting and high read accuracy.
Solution Approach 2:
The invention changes the parameter of read threshold positioning from fixed edge/intermediate values to dynamically calculated medians based on sampled distribution data. By adjusting the threshold parameter to match the actual median of each programming level distribution, the system achieves high read accuracy without excessive complexity.
2Measurement precision
If multiple read operations are performed to accurately estimate distribution medians, then read threshold calculation accuracy improves, but time consumption increases
Solution Approach 1:
The system performs a limited number of preliminary read operations (partial action) using intermediate read thresholds to estimate distribution medians. This partial action is sufficient to achieve accurate threshold calculation without requiring exhaustive sampling, thus balancing measurement precision with time consumption.
Solution Approach 2:
The system uses feedback from preliminary read operations to estimate distribution medians and calculate optimal read thresholds. This feedback mechanism allows the system to improve measurement precision through iterative refinement while controlling the total time required by using the feedback to converge on accurate thresholds efficiently.
3Reliability
If read thresholds are positioned within programming level distributions rather than at edges, then read error rate decreases, but the complexity of determining optimal threshold positions increases
Solution Approach 1:
The system uses self-service by having the memory device itself perform preliminary read operations to estimate its own distribution medians and determine optimal read thresholds. This self-service approach eliminates the need for external complex measurement equipment while achieving high data retrieval accuracy through internally generated distribution characteristics.
Solution Approach 2:
The invention replaces complex mechanical/electrical threshold determination mechanisms with a computational approach using statistical median estimation. By substituting the complex task of physically determining optimal threshold positions with a computational calculation based on sampled data, the system reduces the difficulty of threshold position determination while maintaining high reliability.
Data Source
AI summary
A method for data storage includes storing data in a group of analog memory cells by writing into the memory cells in the group respective storage values, which program each of the analog memory cells to a respective programming state selected from a predefined set of programming states. The programming states include at least first and second programming states, which are applied respectively to first and second subsets of the memory cells, whereby the storage values held in the memory cells in the first and second subsets are distributed in accordance with respective first and second distributions. Respective first and second medians of the first and second distributions are estimated, and a read threshold is calculated based on the first and second medians. The data is retrieved from the analog memory cells in the group by reading the storage values using the calculated read threshold.