NAND Flash RAID Decoding Using Estimated Codewords and Soft Information
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current NAND flash memory systems face challenges in decoding operations due to errors caused by NAND noise and interference, leading to increased latency and degradation in soft decoding capabilities, especially during copy back operations and in RAID encoding/decoding schemes where multiple codewords fail to decode successfully.
Innovation Solution
The method involves generating soft information based on hard decoding results without performing soft sampling, by creating an estimated codeword from the hard decoding of other codewords in the RAID system and combining it with the target codeword to improve error correction capabilities and reduce latency, using a new candidate codeword with 2 bits LLR for soft decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If soft sampling is performed to generate soft information for decoding, then error correction capability is improved, but latency increases and decoding reliability deteriorates under certain stress conditions
Solution Approach 1:
The system performs preliminary hard decoding on multiple codewords before the actual soft decoding is needed. The hard decoding results are stored and reused during soft decoding operations, eliminating the need to perform hard decoding repeatedly and reducing overall latency
Solution Approach 2:
The patent creates copies of hard decoding results from multiple codewords and uses these copies to generate soft information. Instead of performing soft sampling on the original data multiple times, the system copies and combines hard decoding results from different codewords to construct soft information, reducing time consumption
2Reliability
If multiple hard decoding operations are performed on RAID codewords to generate soft information, then soft decoding reliability is improved, but device complexity increases
Solution Approach 1:
The patent divides the decoding process into separate stages: hard decoding of individual codewords is performed first, then the results are combined to generate soft information for soft decoding. This segmentation allows each stage to be optimized independently and simplifies the overall system architecture
Solution Approach 2:
The hard decoding results from multiple codewords serve multiple purposes: they are used both for direct error correction and as input for generating soft information. This multi-functionality reduces the need for separate processing paths and simplifies the decoding system
Data Source
AI summary
A flash memory system may include a flash memory and a circuit for decoding a result of a read operation on the flash memory using a first codeword. The circuit may be configured to generate an estimated codeword based on a result of hard decoding the first codeword and a result of hard decoding a second codeword. The circuit may be further configured to generate soft information based on the hard decoding result of the first codeword and the estimated codeword. The circuit may be further configured to decode the result of the read operation on the flash memory using the soft information.