Flash Memory Read Parameter Tuning After Decoding Failure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solid state storage systems face challenges in error recovery due to insufficient information from single-read operations, leading to inefficiencies in decoding and increased bit error rates, particularly in flash memory devices where hard decisions are not sufficient for data recovery.
Innovation Solution
The method involves processing multiple read values to determine read parameters such as log likelihood ratios (LLRs) and center reference voltages, using signal counts from regions to customize read parameters for error correction, thereby improving decoding accuracy and reducing bit errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If single-read operations are used for data retrieval, then read speed is improved, but decoding accuracy deteriorates due to insufficient information
Solution Approach 1:
The system performs preliminary multiple read operations to collect sufficient signal information before decoding. By reading the same data multiple times and accumulating signal counts, the system prepares adequate soft information in advance, enabling accurate decoding without requiring re-reads during error recovery.
Solution Approach 2:
The patent introduces signal count metrics and soft information as intermediaries between the raw read signals and the decoder. These intermediate representations accumulate information from multiple reads, providing the decoder with more reliable data than single hard decisions while maintaining efficient processing.
2Measurement precision
If multiple read values are processed to determine read parameters, then decoding accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent extracts only the essential signal count information from multiple read operations, focusing on the most relevant metrics for decoding accuracy. By taking out only the necessary soft information and signal counts rather than processing all raw read data, the system maintains decoding accuracy while reducing processing complexity.
Solution Approach 2:
The system transforms multiple read values into standardized read parameters such as signal counts and soft information metrics. This parameter transformation simplifies the data structure and makes it more suitable for decoder input, reducing processing complexity while preserving decoding accuracy.
3Speed
If hard decisions are used for data recovery, then processing speed is improved, but error recovery performance deteriorates
Solution Approach 1:
The patent implements a dynamic approach where the system adapts between hard decisions and soft information processing based on the specific error conditions. For simple cases, fast hard decisions are used; for error-prone cases, the system dynamically switches to using accumulated soft information from multiple reads, optimizing both speed and reliability.
Solution Approach 2:
The system uses feedback from decoding attempts to determine whether hard decisions or soft information should be used. When decoding fails with hard decisions, the feedback triggers a switch to using soft information from multiple reads, improving error recovery performance while maintaining processing efficiency through conditional logic.
Data Source
AI summary
Read parameter estimation techniques are provided that obtain information from multiple read operations to customize read parameters for data recovery. One method comprises performing the following steps, in response to a decoding failure of a codeword of the memory: obtaining at least three read values of the codeword; calculating a signal counts metric value from the at least three reads; computing an optimal reference voltage offset from the signal counts metric and a correlation between optimal reference voltage offsets and a signal counts metric associated with the memory; determining a new center read reference voltage based on a current center reference voltage and the optimal reference voltage offset and performing at least one subsequent read of the codeword following the decoding failure utilizing the new center read reference voltage.