Data Encoding for Non-Volatile Memory Error and Power Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile memory devices face increased bit error rates and inefficiencies in data storage due to cross-coupling effects and power consumption, particularly when storing data patterns with varying threshold voltages, which affects storage density and endurance.
Innovation Solution
Applying a series of transformations including Burrows-Wheeler permutation sorting, move-to-front transformation, and weight-based encoding to reduce entropy and enhance encoding efficiency, thereby improving storage efficiency, reducing programming time, and extending memory endurance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If more bits are stored per cell to increase storage density, then storage density is improved, but bit error rate increases
Solution Approach 1:
The patent applies preprocessing transformations (Burrows-Wheeler permutation sorting, move-to-front transformation) to the data before encoding and storage. These transformations are performed in advance to reduce entropy and optimize the data pattern, thereby improving reliability without sacrificing storage density.
2Reliability
If more parity bits are used to increase error correction capacity, then error correction capacity is improved, but number of bits required to store encoded data increases
Solution Approach 1:
The patent changes the parameters of the data representation by applying entropy-reducing transformations before encoding. This modifies the statistical properties of the data, allowing for more efficient encoding that requires fewer parity bits while maintaining the same error correction capacity.
3Adaptability or versatility
If data patterns with widely differing threshold voltages are stored, then storage flexibility is improved, but cross-coupling effects increase
Solution Approach 1:
The patent applies preliminary transformations to the data that anticipate and counteract the cross-coupling effects. By sorting permutations and applying move-to-front transformation, the data is pre-processed to minimize threshold voltage variations between adjacent cells, thereby reducing cross-coupling effects before storage occurs.
4Productivity
If programming speed is increased to improve productivity, then programming time is reduced, but power consumption during programming increases
Solution Approach 1:
The patent transforms the data parameters to reduce entropy before programming. This preprocessing creates a more favorable data distribution that requires less programming energy per bit, allowing for faster programming speeds with reduced power consumption.
Data Source
AI summary
A data storage device includes a memory and a controller. A method performed in the data storage device includes performing a first transformation of a unit of data to generate a first transformed unit of data. Performing the first transformation includes sorting permutations of the unit of data. The method includes performing a move-to-front transformation of the first transformed unit of data to generate a second transformed unit of data. The method includes performing a weight-based encoding of the second transformed unit of data to generate an encoded unit of data. The encoded unit of data has a same number of bits as the unit of data.