Soft-Decision LLR Generation for Flash Memory LDPC Decoding

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Data storage systems using flash memories face challenges in decoding data due to reduced read margins caused by physical conditions and wear, leading to overlapping voltage distributions, which single-read inputs cannot adequately address, especially when soft-decision data is not provided by SSDs.

Innovation Solution

The implementation of a system and method to generate soft-decision inputs for LDPC decoders using a multiple-read technique, which involves multiple voltage reads to provide more reliable data for error correction and channel estimation, and optimal voltage threshold tracking to enhance decoder performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If single-read input is used for LDPC decoding, then the decoding process is simple and fast, but the reliability of data retrieval deteriorates when voltage distributions overlap due to wear and physical conditions

Engineering Contradiction:

Improvedata retrieval reliabilityVSAvoiddecoding process complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the single-read input into multiple independent read operations at different voltage thresholds. Each read provides a separate binary output that contributes to the overall soft-decision input, dividing the complex reliability problem into manageable discrete measurements that can be combined to improve accuracy

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transitions from single-dimensional hard-decision inputs (binary 0 or 1) to multi-dimensional soft-decision inputs by performing reads at multiple voltage thresholds. This adds dimensional information about the confidence level and probability of each bit value, transforming the input space from simple binary to enriched probabilistic data

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Measurement precision

If multiple voltage reads are performed to generate soft-decision inputs, then the accuracy of error correction improves, but the read time and processing complexity increase

Engineering Contradiction:

Improveerror correction accuracyVSAvoidread time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary reads at multiple voltage thresholds before the actual decoding process. These preliminary measurements establish the soft-decision inputs in advance, allowing the decoder to use pre-computed probability information rather than performing complex calculations during the critical data retrieval phase

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the voltage threshold parameter across multiple read operations. By varying this critical parameter and collecting results at different threshold levels, the system extracts more information about the underlying bit values and their confidence levels, improving measurement precision through parameter variation

Inventive Principle:

Principle #35Parameter changes

3Reliability

If soft-decision inputs are generated from multiple reads, then the number of correctable errors increases, but the processing overhead and computational load increase

Engineering Contradiction:

Improveerror correction capabilityVSAvoidprocessing overhead

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent creates multiple copies of the read data at different voltage thresholds and processes each copy independently to generate separate binary outputs. These copies are then combined to form the soft-decision input, allowing the system to leverage redundant information from multiple measurements without requiring complex real-time processing of raw data

Inventive Principle:

Principle #26Copying

4Adaptability or versatility

If voltage threshold tracking is implemented to optimize reads, then the adaptability to wear and physical conditions improves, but the system complexity and calibration requirements increase

Engineering Contradiction:

Improveadaptability to wear conditionsVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements feedback through voltage threshold tracking that monitors the performance of reads at different thresholds and adjusts future read operations accordingly. This feedback mechanism allows the system to adapt to wear and physical conditions by learning from past measurements and optimizing threshold selection for current operating conditions

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent makes the voltage threshold dynamic rather than fixed. The threshold tracking capability allows the read thresholds to change and adapt based on observed data characteristics and wear patterns, enabling the system to respond dynamically to changing physical conditions in the memory device

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS8924824B1Soft-decision input generation for data storage systems

Publication Date: 2014.12.30 WESTERN DIGITAL TECHNOLOGIES INC

AI summary

An error management system for a data storage device can generate soft-decision log-likelihood ratios (LLRs) using multiple reads of memory locations. 0-to-1 and 1-to-0 bit flip count data provided by multiple reads of reference memory locations can be used to generate probability data that is used to generate possible LLR values for decoding target pages. Possible LLR values are stored in one or more look-up tables.