Flash Memory Reference Voltage Updating for Read Error Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Non-volatile data storage devices, such as flash memory devices, face accuracy issues in reading stored data due to variations in programmed states caused by factors like data retention and program disturb conditions, leading to errors when using fixed reference voltages over time.

Innovation Solution

The system updates a set of reference voltages based on a comparison of error counts, selecting or replacing reference voltages to reduce read errors, conditioned on a target number of error correction code pages read and the difference in mean error counts since the last update, thereby maintaining data accuracy and reliability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If reference voltages are updated frequently to reduce read errors, then data reading accuracy is improved, but processing resources and time are increased

Engineering Contradiction:
Improvedata reading accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system implements periodic updating of reference voltages based on read error counts. Instead of continuous updating, the controller periodically evaluates whether updates are needed by comparing error counts against thresholds, and only updates when necessary. This periodic approach maintains data reading accuracy while minimizing unnecessary processing time and resource consumption.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system uses feedback from read error counts to dynamically control reference voltage updates. The controller monitors error counts during data reading operations and uses this feedback to determine when updates should occur. This feedback mechanism ensures updates happen only when actually needed, optimizing the balance between accuracy and processing efficiency.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If reference voltages are updated based on error counts, then data reading accuracy is improved, but device complexity is increased

Engineering Contradiction:
Improvedata reading accuracyVSAvoidcontrol logic complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system implements dynamic reference voltage updating where the update behavior adapts based on monitored error counts. The controller dynamically adjusts whether to update reference voltages by comparing current error counts against stored thresholds. This dynamic approach improves reading accuracy while keeping the control logic manageable through adaptive rather than static decision-making.

Inventive Principle:
Principle #15Dynamics

3Reliability

If reference voltages are updated to reduce errors, then reliability is improved, but productivity is reduced due to processing overhead

Engineering Contradiction:
Improvedata storage reliabilityVSAvoiddata processing throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system applies partial updating of reference voltages by only updating when error counts exceed specific thresholds. Rather than continuously updating or updating all reference voltages, the controller performs selective updates based on actual error conditions. This partial action approach maintains data storage reliability while minimizing the processing overhead that would otherwise reduce productivity.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS8874992B2Systems and methods to initiate updating of reference voltages
Publication Date: 2014.10.28 SANDISK TECHNOLOGIES LLC
  • US8874992B2 patent drawing
  • US8874992B2 patent drawing
  • US8874992B2 patent drawing

AI summary

In a data storage device that includes a non-volatile memory, a method includes determining that a current error correction code page count (CEC) is at least as large as a target error correction code page count (TEC). The CEC is a page count of error correction code (ECC) pages of data read from the memory during a time period from a previous time to a particular time using a set of reference voltages. In response to the CEC being at least as large as the TEC, the method includes updating a subset of the set of reference voltages conditioned upon a difference between a current mean error count (CMEC) and a previous mean error count being at least as large as a target mean delta error. The CMEC is based on a count of read errors associated with the ECC pages read during the time period.