Read Level Threshold Estimation Using Histogram Data Structures

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional memory sub-systems rely on manufacturing default read level thresholds, which are not adjusted over the life of the system, leading to increased errors and reduced lifespan due to shifting programming distributions and resource-intensive voltage sweeping methods for determining ideal read level thresholds.

Innovation Solution

The method involves performing sampling operations at various read level thresholds, generating a histogram to estimate the error counts, and selecting a read level threshold with the lowest error count to improve data integrity and extend the memory sub-system's lifespan by dynamically adjusting read level thresholds.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If voltage sweeping methods are used to determine ideal read level thresholds, then measurement precision is improved, but use of energy and time are significantly increased

Engineering Contradiction:

Improveread level threshold accuracyVSAvoidthreshold determination time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs sampling operations at various read level thresholds during initial programming operations to pre-determine optimal thresholds before they are needed for actual data retrieval. This preliminary characterization eliminates the need for time-consuming voltage sweeping during normal operation, as the histogram data structure is built in advance during programming cycles.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of performing exhaustive voltage sweeping to measure the entire programming distribution, the patent creates a histogram data structure that copies and represents the essential characteristics of the distribution through sampling. This histogram serves as a simplified model that enables rapid threshold determination without replicating the full measurement process.

Inventive Principle:

Principle #26Copying

2Measurement precision

If voltage sweeping methods are used to determine ideal read level thresholds, then measurement precision is improved, but computational resources are significantly consumed

Engineering Contradiction:

Improveread level threshold accuracyVSAvoidcomputational energy for threshold estimation

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent performs sampling operations at selected read level thresholds rather than conducting exhaustive voltage sweeping across the entire voltage range. By sampling at specific points and using histogram analysis to estimate the programming distribution characteristics, the system achieves sufficient measurement precision with substantially reduced computational energy consumption.

Inventive Principle:

Principle #16Partial or excessive action

3Device complexity

If manufacturing default read level thresholds are used, then device complexity is reduced, but reliability deteriorates due to shifting programming distributions

Engineering Contradiction:

Improvethreshold management simplicityVSAvoiddata retrieval accuracy

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The patent implements dynamic adjustment of read level thresholds by maintaining histogram data structures that are updated based on programming operations. The system transitions from static manufacturing default thresholds to dynamic thresholds that adapt to shifting programming distributions, improving reliability while managing complexity through efficient histogram maintenance and estimation algorithms.

Inventive Principle:

Principle #15Dynamics

4Measurement precision

If extensive sampling is performed to determine read level thresholds, then measurement precision is improved, but loss of time and resources are increased

Engineering Contradiction:

Improvethreshold estimation accuracyVSAvoidsystem operation efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent performs sampling operations during initial programming cycles to pre-establish histogram data structures that capture programming distribution characteristics. This preliminary sampling eliminates the need for extensive measurement operations during normal read operations, thereby improving overall system productivity while maintaining accurate threshold estimation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The histogram data structure is continuously updated and maintained based on programming operations, allowing the system to leverage existing programming activity for threshold characterization. This continuous updating approach ensures that threshold estimates remain accurate without requiring separate extensive sampling campaigns, thereby maintaining high system productivity.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS11789640B2Estimation of read level thresholds using a data structure

Publication Date: 2023.10.17 MICRON TECHNOLOGY INC

AI summary

A data structure that identifies a characteristic of a region that is located between programming distributions of the memory device and that corresponds to read level thresholds at the region is determined. An estimator type is selected from a plurality of estimator types corresponding with the data structure. A read level threshold of the read level thresholds is estimated using the selected estimator type. A read operation is performed at the memory device using the read level threshold estimated using the selected estimator type.