Flash Memory Noise Adaptation via Dynamic Program-Verify Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Noise in the common source line of a flash memory device can cause errors during read and program-verify operations by altering the threshold voltage distribution of memory cells, leading to incorrect identification of on-cells as off-cells.
Innovation Solution
A method and device that detect the noise level of the common source line and adjust the number of program-verify or read operations based on this level, performing multiple operations if the noise exceeds a threshold and fewer operations if it does not, to maintain accurate cell identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of program-verify operations is increased to reduce errors caused by noise, then the reliability of data retrieval is improved, but the time required for programming operations increases
Solution Approach 1:
The patent applies dynamics by making the number of program-verify operations variable rather than fixed. The control circuit dynamically adjusts the number of verify operations based on real-time noise level detection in the common source line, performing more verify operations when noise is high and fewer when noise is low, thus optimizing both reliability and time efficiency
Solution Approach 2:
The patent changes the parameter of operation frequency based on noise conditions. By detecting noise levels and adjusting the number of program-verify operations accordingly, the system adapts its operational parameters to maintain accurate cell identification while minimizing unnecessary operations and time consumption
2Reliability
If multiple read operations are performed to compensate for noise-induced errors, then the accuracy of data retrieval is improved, but the operational speed decreases
Solution Approach 1:
The patent makes the read operation frequency dynamic by adjusting it based on detected noise levels. When noise is detected in the common source line, the system performs multiple read operations to ensure accuracy; when noise is low, it performs fewer operations, thus maintaining speed while ensuring reliability
Solution Approach 2:
The system changes the operational parameter of read frequency based on noise conditions. By detecting noise levels and adapting the number of read operations, the patent optimizes the balance between retrieval accuracy and operational speed
3Reliability
If the number of program-verify operations is increased to ensure accurate cell identification, then the reliability is improved, but the productivity of the memory device decreases
Solution Approach 1:
The patent applies dynamics by adjusting the number of program-verify operations based on real-time noise detection. When noise levels are high, more verify operations are performed to ensure programming accuracy; when noise is low, fewer operations are performed, thus maintaining productivity while ensuring reliability
Solution Approach 2:
The system changes the programming operation parameters based on noise conditions. By detecting noise levels and adapting the number of verify operations, the patent optimizes the balance between programming accuracy and throughput, preventing unnecessary operations that would reduce productivity
Data Source
AI summary
In one embodiment, the method includes detecting a noise level of a common source line, and adjusting a frequency of program-verify operations on a memory cell during a programming loop based on the detected noise level.