NAND Memory Read Voltage Selection for LDPC Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As NAND memory integration and bit density increase, bit error rates rise, leading to reliability issues, and existing error correction methods like LDPC require multiple iterative calculations, wasting time and causing errors.
Innovation Solution
A method for operating memory that involves reading memory cells with multiple read voltages to determine the valley position of threshold voltage distributions, reducing the need for iterative calculations and enhancing error correction capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LDPC technology is employed to enhance error correction capability, then data reliability is improved, but read time increases due to multiple iterative calculations
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing optimal read voltages in a lookup table before actual data reading operations. When reading data, the system directly retrieves the pre-determined optimal read voltage based on the specific data state, eliminating the need for time-consuming iterative calculations during the read operation itself. This preliminary preparation resolves the contradiction by maintaining high data reliability through accurate voltage selection while significantly reducing read time.
2Reliability
If multiple iterative calculations are performed for error correction, then error correction capability is improved, but calculation complexity increases
Solution Approach 1:
The patent applies the copying principle by creating a lookup table that stores pre-computed optimal read voltages for different data states. Instead of performing complex iterative calculations during each read operation, the system copies the pre-determined voltage values from the lookup table for immediate use. This approach maintains error correction capability while dramatically reducing calculation complexity and processing time during actual data retrieval operations.
3Ease of operation
If read voltage is not optimized, then read operation is simple, but error rate increases
Solution Approach 1:
The patent applies parameter changes by dynamically selecting different read voltages based on the specific data state being read. The system determines the current data state and retrieves the corresponding optimal read voltage from the lookup table, ensuring that the read operation uses the most appropriate voltage parameter for that particular state. This resolves the contradiction by maintaining operational simplicity through automated voltage selection while significantly reducing error rates through optimized voltage parameters.
Data Source
AI summary
A memory includes an array of memory cells and a peripheral circuit. A memory cell is configured to store N bits of data, and a plurality of memory cells has 2N data states. The peripheral circuit is configured to: read the array of memory cells with a first read voltage and determine a first number of memory cells with a threshold voltage greater than the first read voltage that is greater than a first initial threshold voltage; read the array of memory cells with a second read voltage and determine a second number of memory cells with a threshold voltage greater than the second read voltage that is greater than a second initial threshold voltage and less than the first initial threshold voltage; determine a valley position of distribution interval of threshold voltages for a 2N-th data state based on a difference between the second number and the first number.