Memory ECC Recovery Using Soft Reads Across Data Groups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As data transfer speeds increase and channels become more complex, existing error correction methods struggle to effectively handle data bit errors, necessitating improved error encoding and decoding techniques for reliable data restoration.
Innovation Solution
The proposed method involves reading data from multiple data groups in a memory system, performing multiple error correction code (ECC) decodings, and using soft read values to determine the reliability of data bits, allowing for correction of errors through XOR operations and ECC decodings, ensuring accurate data recovery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple ECC decodings and soft read operations are performed to improve error correction reliability, then data integrity is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary ECC decoding attempts before final data restoration. Multiple ECC decoding operations are attempted in sequence, with soft read values obtained in advance to assess error patterns. This preliminary action allows the system to identify correctable errors early and only perform additional complex operations when necessary, reducing overall processing time while maintaining high reliability.
Solution Approach 2:
Soft read values serve as an intermediary mechanism between the raw read data and final error correction. The system obtains soft read values that provide reliability information about each bit, acting as a mediator to guide the ECC decoding process. This intermediary layer enables intelligent decision-making about which bits require correction and which can be trusted, improving efficiency without sacrificing accuracy.
2Reliability
If multiple ECC decodings and soft read operations are performed to improve error correction reliability, then data integrity is improved, but computational complexity increases
Solution Approach 1:
The patent applies partial action by performing ECC decoding only on the necessary portions of data. Instead of uniformly applying complex decoding to all data, the system uses soft read values to identify specific bits or data groups that require correction. This selective approach reduces computational complexity by focusing resources only where errors are detected, while still achieving high data integrity through targeted correction.
3Measurement precision
If data is re-read multiple times with different voltages to obtain soft read values, then error detection accuracy is improved, but read operation time increases
Solution Approach 1:
The patent uses periodic action by performing read operations at different voltage levels in a structured sequence. Multiple reads are conducted with varying read voltages to obtain diverse soft read values, which are then combined to improve error detection accuracy. This periodic multi-voltage reading approach systematically gathers reliability information without continuously increasing read time, as the voltage variations are applied in a controlled, periodic manner.
Data Source
AI summary
An operating method of a memory system includes: reading a first data from a particular data group among a plurality of data groups included in a memory device; performing a first error correction code (ECC) decoding for the first data; when the first ECC decoding fails, reading a plurality of the remaining data other than the first data from the particular data group; performing a second ECC decoding for the plurality of the remaining data; when the second ECC decoding fails, identifying data, to which the second ECC decoding fails, among the plurality of the remaining data; obtaining first and second soft read values respectively corresponding to the first data, to which the first ECC decoding fails, and the second data, to which the second ECC decoding fails; determining reliability of the first and second data based on the first and second soft read values; and correcting the first data based on the reliability of the first and second data.