ECC Memory Read Retry With Erasure-Aware BCH Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Nonvolatile memory devices, such as NAND Flash memories, experience increasing reading/writing error rates due to degradation over time, exceeding the correction capacity of embedded error-correcting codes, leading to reliability issues.
Innovation Solution
A method and device that utilize an erasure-locator module to perform additional readings of memory locations, generate a modified string with erasure indicators, and execute subsequent decoding attempts to correct errors beyond the initial correction capacity of cyclic error-correcting codes like BCH codes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If error-correcting codes are embedded in memory devices to correct reading/writing errors, then reliability is improved, but the correction capacity is limited and cannot handle errors exceeding a certain threshold
Solution Approach 1:
The patent performs preliminary actions by conducting multiple readings of the same memory location before decoding. This preliminary data collection allows the system to identify patterns and detect erasures that single readings would miss, enabling the ECC to handle errors beyond its traditional correction capacity.
Solution Approach 2:
The patent adds a temporal dimension to error detection by performing readings at different times (multiple readings). This transforms the problem from a single-point-in-time error detection to a multi-point-in-time analysis, allowing the system to distinguish between random errors and systematic erasures.
2Reliability
If multiple readings are performed to detect erasures, then reliability is improved, but reading time increases
Solution Approach 1:
Multiple readings are performed as a preliminary step before decoding to collect sufficient data for reliable erasure detection. This preliminary data gathering enables the system to make informed decisions about which bits are erasures, improving overall decoding success.
Solution Approach 2:
The system uses the memory device itself to provide the additional readings needed for erasure detection, rather than requiring external intervention or complex additional hardware. The same read circuitry is reused multiple times, making the process self-contained and efficient.
3Manufacturing precision
If erasure location is performed through multiple readings and comparison, then manufacturing precision is improved in terms of error identification, but device complexity increases
Solution Approach 1:
The decoding process is segmented into distinct phases: first performing multiple readings, then comparing results to identify potential erasures, and finally performing decoding with identified erasure locations. This segmentation makes the complex process more manageable and systematic.
Solution Approach 2:
The system uses feedback from multiple readings to identify erasure locations, which then informs the decoding process. This feedback mechanism allows the system to adapt its decoding strategy based on the actual data quality observed, improving error identification accuracy.
Data Source
AI summary
A reading method for a memory device with error-correcting encoding envisages the steps of: carrying out a first reading of a plurality of memory locations (A0, A1, . . . , ALS−1) to generate a first recovered string (S1), and performing a first decoding attempt using the first recovered string (S1). When the first decoding attempt fails, the memory locations are read at least one second time, and at least one second recovered string (S2-SN) is generated. On the basis of a comparison between the first recovered string (S1) and the second recovered string (S2-SN), a modified string (SM) is generated, in which erasures (X) are located, and at least one second decoding attempt is carried out using the modified string (SM).