Non-Volatile Memory ECC Recovery Across Defective Storage Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As semiconductor memory devices become more integrated, the likelihood of defective memory cells increases, leading to decreased yield and reliability, with existing error correction methods being inadequate in handling simultaneous defects across multiple storage regions.
Innovation Solution
A method for controlling error check and correction (ECC) in non-volatile memory devices involves storing write data in multiple regions, performing ECC encoding, individual ECC decoding, and providing logic operation data through bit-by-bit OR or AND operations when individual decoding fails, followed by combined ECC decoding based on the logic operation data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the degree of integration increases to increase storage capacity, then storage capacity is improved, but the likelihood of defective memory cells increases leading to decreased yield and reliability
Solution Approach 1:
The patent divides the data storage and retrieval process into multiple independent regions, where data is stored across several separate memory regions and retrieved through individual read operations. This segmentation allows the system to handle defects in one region independently without affecting other regions, thereby maintaining reliability while achieving high storage capacity through increased integration.
Solution Approach 2:
The patent combines multiple individual read data from different memory regions through logical operations (OR/AND operations) to reconstruct the original data. By merging results from multiple regions, the system achieves both high storage capacity and improved reliability, as the combination process can correct errors that occur in individual regions.
2Reliability
If redundant memory cells are used to repair defective memory cells, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent creates multiple copies of data across different memory regions, where each region stores the same encoded data independently. Instead of using complex redundant memory cell structures, the system uses simple data copying to multiple regions, reducing device complexity while maintaining the ability to repair defective cells through majority voting or logical operations on the copies.
Solution Approach 2:
The patent changes the approach from physical redundancy (additional memory cells) to operational redundancy (multiple read operations and logical combinations). By altering the parameters of the read operation rather than the physical structure, the system achieves defect repair capability without increasing device complexity.
3Reliability
If ECC encoding is performed on write data and stored in multiple regions, then error correction capability is improved, but processing time increases
Solution Approach 1:
The patent performs ECC encoding on data before writing to multiple regions, preparing the data in advance for potential error correction. This preliminary action allows the system to store error-correcting codes alongside the data in each region, so that when read operations occur, the correction process is already prepared and can be executed quickly without significant time loss.
Solution Approach 2:
The patent applies ECC encoding to only the necessary portions of data that will be stored in multiple regions, rather than encoding all data uniformly. This partial action approach reduces overall processing time while still providing error correction capability where it is most needed - in the multi-region storage and retrieval operations.
4Reliability
If individual ECC decoding is performed on each read data from multiple regions, then error detection capability is improved, but processing complexity increases
Solution Approach 1:
The patent divides the decoding process into separate individual ECC decoding operations for each memory region, with each decoder handling only its own region's data independently. This segmentation of the decoding process reduces the complexity of each individual decoder while maintaining strong error detection capability, as each segment can detect errors in its own region without being burdened by the complexity of analyzing all regions simultaneously.
Solution Approach 2:
The patent uses the same ECC decoding algorithm and logic operation framework across all memory regions, making the decoding process universal and multi-functional. This universality allows the system to detect errors in any region using the same standardized approach, reducing overall processing complexity through reuse of proven decoding logic rather than requiring custom complex decoders for each region.
Data Source
AI summary
A method for controlling error check and correction (ECC) of a non-volatile memory device includes storing write data in a plurality of storing regions. The write data may be generated by performing ECC encoding. Individual ECC decoding may be performed based on each of a plurality of read data read out from the storing regions. Logic operation data may be provided by performing a logic operation of the read data when the individual ECC decoding fails with respect to all of the read data. Combined ECC decoding may be performed based on the logic operation data.