Memory Read Retry Using Syndrome Weights for Soft Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Solid state storage devices face challenges in maintaining accurate read threshold voltages over time due to memory cell leakage and disturbances, leading to data errors and the need for improved read retry operations.
Innovation Solution
The method involves estimating written data based on syndrome weights by reading codewords multiple times with varying reference voltages, calculating cell voltage probability distributions, and dynamically selecting log likelihood ratio values to optimize read reference voltages for improved decoding accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If read retry operations are performed multiple times with varying read voltages, then decoding accuracy is improved, but read time and operational complexity increase
Solution Approach 1:
The patent implements feedback by using syndrome weight measurements from each read attempt to dynamically adjust subsequent read operations. The controller measures the syndrome weight, compares it against thresholds, and uses this feedback to determine whether to perform additional read retries with adjusted voltages, thereby optimizing the balance between decoding accuracy and read time.
Solution Approach 2:
The read retry mechanism is made dynamic by adjusting read voltages and deciding whether to retry based on real-time syndrome weight measurements. Instead of fixed retry schedules, the system adaptively modifies read parameters during operation, allowing it to terminate early when sufficient accuracy is achieved while performing multiple retries only when necessary.
2Productivity
If syndrome weight measurement is used to determine read retry, then read operations are optimized, but measurement and control complexity increase
Solution Approach 1:
The patent introduces syndrome weight as an intermediary metric that bridges the gap between raw read data and decoding decisions. Instead of directly analyzing complex error patterns, the system uses syndrome weight as a simplified intermediate measurement that captures error severity, making control decisions more manageable while still optimizing read operations.
Solution Approach 2:
The system changes the parameter being measured from raw bit error patterns to syndrome weight, which is a transformed parameter that more efficiently represents error conditions. This parameter transformation simplifies the control logic while maintaining the ability to make informed decisions about read retries.
3Reliability
If multiple reads with varying reference voltages are performed, then soft decision quality is enhanced, but the number of read operations and energy consumption increase
Solution Approach 1:
The patent applies partial action by performing only the number of reads necessary to achieve sufficient soft decision quality, as determined by syndrome weight thresholds. Instead of always performing a fixed number of reads with varying voltages, the system performs reads partially—stopping early when the syndrome weight indicates adequate quality has been achieved, thereby reducing unnecessary energy consumption.
4Reliability
If read threshold voltages are adjusted dynamically, then data retrieval reliability is improved, but device complexity and operational difficulty increase
Solution Approach 1:
The system implements self-service by automatically adjusting read threshold voltages based on syndrome weight measurements without requiring external intervention or complex manual calibration. The controller autonomously monitors syndrome weights and modifies read parameters accordingly, making the system self-regulating while maintaining high reliability.
Data Source
AI summary
Methods and apparatus are provided for read retry operations that estimate written data based on syndrome weights. One method comprises reading a codeword from a memory multiple times using multiple read reference voltages; obtaining a syndrome weight for each of the readings of the codeword; identifying a given reading of the codeword having a substantially minimum syndrome weight; and estimating a written value of the codeword based on the given reading. Two cell voltage probability distributions of cell voltages are optionally calculated for each possible cell state of the memory, based on the estimated written value and plurality of readings of the codeword. The cell voltage probability distributions are used to (i) dynamically select log likelihood ratio values for a failing page, (ii) determine a read reference voltage that gives a desired log likelihood ratio value, or (iii) dynamically select log likelihood ratio values for the page populations associated with the distributions.