NAND Data Encoding Schemes for Lower Programmed 0 Density
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
NAND memory cell endurance is degraded due to the high number of logical '0's programmed during each program/erase cycle, leading to reduced memory cell lifespan and storage capacity.
Innovation Solution
The proposed data encoding and decoding schemes convert user data into a format with a reduced number of logical '0's, thereby minimizing the stress on memory cells during programming cycles, while maintaining a high degree of variation in stored bit patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If user data is programmed directly to NAND memory cells, then storage capacity is maximized, but memory cell endurance is degraded due to high number of logical '0's programmed
Solution Approach 1:
The patent applies parameter changes by transforming the data representation parameters through encoding schemes that alter the distribution of logical '0' and '1' bits. Specifically, it uses techniques such as inverting least significant bits, applying Gray codes, or using other encoding transformations that change the bit pattern parameters to reduce the frequency of logical '0's, thereby improving memory cell endurance while preserving storage capacity
Solution Approach 2:
The patent introduces an intermediary encoding layer between the host system and the NAND memory cells. This intermediary transformation layer processes user data through encoding schemes that modify the bit patterns before storage, reducing the direct impact of logical '0's on memory cells while maintaining the ability to store diverse data patterns
2Reliability
If encoding schemes are applied to reduce logical '0's, then memory cell endurance is improved, but data encoding complexity increases
Solution Approach 1:
The patent segments the encoding process into simple, modular operations that can be applied independently to different portions of data. The encoding scheme is broken down into basic transformations such as bit inversion, bit shifting, or lookup table applications, which can be implemented through simple logic circuits or software functions, thereby limiting the increase in encoding complexity
Solution Approach 2:
The patent employs lightweight encoding transformations that require minimal computational resources and can be executed quickly. The encoding schemes use simple operations such as XOR transformations, bit inversions, or pre-computed lookup tables that can be applied with minimal processing overhead, keeping the complexity increase manageable
Data Source
AI summary
A method includes receiving user data having a number of first bits. The method further includes encoding the user data by generating a number of second encoded bits having a first quantity of bits greater than that of the number of first bits. The number of second encoded bits can include one or more bits having a particular binary value and a quantity of the one or more bits is less than a threshold quantity. The method further includes writing the number of second encoded bits as the user data to a memory.