Flash Memory Soft-Data Decoding for Lower Read Latency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Flash memory devices experience significant latency due to the need for multiple comparisons to determine binary values from multi-level cells, which affects data storage and access efficiency.

Innovation Solution

The implementation of a system that includes a data read circuit, first and second data decoding circuits, and encoding/decoding circuits using low density parity check (LDPC) codewords and voltage value encoding, allowing for efficient storage and retrieval of data by comparing voltages with multiple thresholds to generate soft data values, which are then decoded to yield data outputs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If multiple voltage comparisons are performed to determine binary values from multi-level cells, then data density is increased, but access latency increases

Engineering Contradiction:

Improvedata densityVSAvoidaccess latency

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The patent performs preliminary voltage comparisons during the read operation to generate soft data values that indicate the likelihood of each binary value. By pre-computing these probability indicators and using them in subsequent decoding operations, the system reduces the need for repeated comparisons and iterative error correction, thereby reducing access latency while maintaining high data density

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the parameter representation from hard binary decisions to soft probability values. Instead of immediately determining binary values through multiple threshold comparisons, the system generates soft data values that represent the probability distribution of possible binary values. This parameter transformation enables more efficient error correction and reduces the number of comparison operations needed, thus reducing latency while preserving data density

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If multiple voltage comparisons are performed to determine binary values, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improvebinary value determination accuracyVSAvoiddecoding circuit complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces soft data values as an intermediary representation between the voltage readings and the final binary decisions. These soft data values serve as probability indicators that mediate the conversion from analog voltage levels to digital binary values. This intermediary layer simplifies the decoding process by providing probabilistic information that can be processed through efficient algorithms, reducing the complexity of the decoding circuit while maintaining high measurement precision

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces complex mechanical comparison circuits with software-based decoding algorithms that process soft data values. Instead of using multiple hardware comparators and complex logic circuits to determine binary values, the system uses programmable decoders that can efficiently process probability indicators. This substitution of mechanical/hardware complexity with algorithmic processing reduces device complexity while maintaining or improving measurement precision

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10108489B2Systems and methods for efficient soft data based flash memory data recovery

Publication Date: 2018.10.23 AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE LTD

AI summary

Embodiments are related to systems and methods for data storage, and more particularly to systems and methods for storing and accessing data from a flash memory.