Flash Memory Program State Determination via Data Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing flash memory devices face inefficiencies in determining program states, particularly as the number of bits stored increases, due to the reliance on flag cells and the complexity of threshold voltage distributions.
Innovation Solution
Implementing a data value distribution function to produce transformed data with a predetermined distribution, allowing for the determination of program states based on the proportion of cells meeting specific criteria, such as threshold voltages, without relying solely on flag cells, and using a program state determiner circuit to control read and write operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If flag cells are used to store program state information, then program state can be tracked, but device complexity increases and scalability to multi-bit storage is limited
Solution Approach 1:
The invention extracts the program state information directly from the data cells themselves by analyzing threshold voltage distributions, rather than storing it separately in flag cells. This removes the need for dedicated flag cell structures while maintaining program state tracking capability.
Solution Approach 2:
The data cells serve dual functions: storing user data and indicating program state through their threshold voltage distributions. This eliminates the need for separate flag cells, as the same cells provide both data storage and status indication functions.
2Quantity of substance
If the number of bits stored per cell increases, then storage capacity improves, but the complexity of determining program state increases due to multiple cell states
Solution Approach 1:
The invention replaces complex sequential state determination mechanisms with statistical analysis of threshold voltage distributions. Instead of tracking individual cell transitions through multiple states, the system uses distribution characteristics (mean, variance) to determine overall program state, simplifying the determination process for multi-bit storage.
Solution Approach 2:
The invention changes the approach from discrete state tracking to continuous parameter analysis. By analyzing threshold voltage distribution parameters (mean voltage, variance) rather than discrete cell states, the system can efficiently determine program state for multi-bit storage without increasing complexity.
3Reliability
If traditional threshold voltage distribution analysis is used, then program state can be determined, but the process is time-consuming and reduces operational efficiency
Solution Approach 1:
The invention performs partial analysis by sampling a subset of cells to determine distribution characteristics rather than analyzing every cell. This statistical sampling approach provides sufficient program state information while significantly reducing the time required compared to exhaustive analysis.
Solution Approach 2:
The system performs preliminary analysis of threshold voltage distributions during programming operations to determine when programming is complete, rather than waiting for all cells to be fully programmed. This allows earlier termination of programming operations and improves overall operational efficiency.
Data Source
AI summary
In a memory device, a proportion of at least one cell state in a unit of the memory is determined. A program state of the unit of the memory is determined based on the determined proportion of the at least one cell state. Determining a proportion of at least one cell state in a unit of the memory may be preceded by processing data to be stored in the unit of the memory according to a data value distribution function to produce transformed data having data values conforming to a predetermined distribution and storing the transformed data in the unit of the memory. The distribution function may be configured, for example, to provide a uniform distribution of data values in the unit of the memory.