Memory Error Correction Circuit Blocks Multi-Bit Write-Back

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Semiconductor memory devices face errors during data writing, reading, or storage, leading to single-bit or multi-bit errors, which can be incorrectly corrected by error correction circuits, resulting in error accumulation.

Innovation Solution

A method and system that includes a memory controller and device with an error correction circuit that detects errors, corrects single-bit errors, and blocks write-back of corrected data when multi-bit errors are detected to prevent further errors during the scrubbing operation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If error correction circuit corrects all detected errors, then data accuracy improves, but multi-bit errors are incorrectly corrected causing error accumulation

Engineering Contradiction:

Improvedata accuracyVSAvoiderror accumulation

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The error correction circuit changes its correction behavior based on the error type parameter. When a multi-bit error is detected, the circuit changes from correcting errors to blocking write-back operations, preventing incorrect corrections while maintaining accurate correction for single-bit errors

Inventive Principle:

Principle #35Parameter changes

2Reliability

If error correction is performed on all errors, then data integrity improves, but incorrect correction of multi-bit errors occurs

Engineering Contradiction:

Improvedata integrityVSAvoidcorrection accuracy

Core Design Contradiction:

ReliabilityVSManufacturing precision

Solution Approach 1:

The system changes the correction parameter based on error detection results. For multi-bit errors, the write-back block parameter is activated to prevent incorrect corrections, while single-bit errors continue to be corrected normally, ensuring correction accuracy is maintained

Inventive Principle:

Principle #35Parameter changes

3Reliability

If scrubbing operation continuously corrects errors, then data reliability improves, but error accumulation occurs due to multi-bit error miscorrection

Engineering Contradiction:

Improvedata reliabilityVSAvoiderror accumulation

Core Design Contradiction:

ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The system converts the harmful effect of multi-bit error detection into a beneficial blocking action. By detecting multi-bit errors and blocking write-back operations, the system prevents incorrect corrections from causing error accumulation, while still maintaining the beneficial scrubbing function for single-bit errors

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentUS11921579B2Method of operating memory device, method of operating memory controller and memory system

Publication Date: 2024.03.05 SAMSUNG ELECTRONICS CO LTD

AI summary

A method of operating a memory device is provided. The method includes: receiving a first command from a controller; activating a page of a memory cell array based on the first command; reading data of the activated page; detecting an error from the read data; correcting the detected error to generate error correction data; writing back the error correction data to the activated page in based on the detected error being a single-bit error; and blocking write-back of the error correction data to the activated page based on the detected error being a multi-bit error.