Nonvolatile Memory Programming Verification via Selective Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Nonvolatile memory devices face inefficiencies in programming operations due to the need for accurate verification of program states, particularly in flash memory devices, where noise on the common source line can lead to incorrect determination of threshold voltages, resulting in incomplete programming and increased verification time.
Innovation Solution
A method involving a first sensing operation to determine if any off-cells exist among target memory cells, followed by a selective second sensing operation only on those cells identified as off-cells, optimizing verification by reducing unnecessary operations and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a verification operation is performed on all memory cells connected to a wordline, then the program state of target memory cells can be determined, but noise on the common source line causes incorrect determination and increases verification time
Solution Approach 1:
The verification process is segmented into two distinct operations: a first sensing operation performed on all memory cells connected to the wordline, and a second sensing operation selectively performed only on target memory cells identified as off-cells during the first operation. This segmentation allows the patent to reduce verification time by avoiding redundant sensing of already-verified cells while maintaining accurate program state determination through the selective second sensing operation.
Solution Approach 2:
The patent applies local quality by performing the second sensing operation with different characteristics than the first sensing operation. Specifically, the second sensing operation is applied locally only to off-cell memory cells that require verification, rather than uniformly to all memory cells. This localized approach reduces the overall verification time while ensuring accurate program state determination for the specific cells that need it.
2Loss of time
If selective sensing is performed only on target memory cells, then verification time is reduced, but accuracy of program state determination deteriorates due to noise on common source line
Solution Approach 1:
The first sensing operation serves as a preliminary action that identifies which memory cells are off-cells before the second sensing operation is performed. This preliminary classification allows the system to focus subsequent verification efforts only on the specific cells that require it, reducing overall verification time while maintaining accuracy through the targeted second sensing operation on identified off-cells.
Solution Approach 2:
The first sensing operation provides feedback information about the state of memory cells, which is then used to determine which cells require the second sensing operation. This feedback mechanism ensures that the second sensing operation is applied selectively to off-cell memory cells, maintaining accurate program state determination while avoiding redundant verification of cells that are already in the desired state.
3Measurement precision
If multiple sensing operations are performed on all memory cells, then program state accuracy is improved, but device complexity and operation time increase
Solution Approach 1:
The verification process is made dynamic by conditionally performing the second sensing operation based on the results of the first sensing operation. Rather than executing a fixed sequence of sensing operations on all cells, the system dynamically determines which cells require the second sensing operation, reducing overall complexity and operation time while maintaining verification accuracy for target memory cells.
Solution Approach 2:
The patent changes the parameters of the sensing operation by applying different sensing schemes to different groups of memory cells. The first sensing operation uses one set of parameters applied to all cells, while the second sensing operation uses different parameters applied selectively to off-cell memory cells. This parameter differentiation reduces overall verification complexity while maintaining accuracy.
Data Source
AI summary
A method of programming a nonvolatile memory device comprises programming target memory cells among a plurality of memory cells connected to a wordline, performing a first sensing operation on the plurality of memory cells, and selectively performing a second sensing operation on the target memory cells based on a result of the first sensing operation.