Multi-Level Cell Programming Method for Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional multi-level cell memory programming requires complex address inputs and lengthy programming times due to the need for single programming loops across both half cells, increasing complexity and time consumption.
Innovation Solution
A method that programs 2n-group data sequentially into multi-level cells using a target address, separating data between left and right half cells within a single programming loop, reducing the complexity of address inputs and shortening programming time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional single programming loop is used for both half cells, then threshold voltage distribution can be maintained, but programming time and address input complexity increase significantly
Solution Approach 1:
The patent segments the programming process into two separate programming loops: a first programming loop for programming the left half cell, and a second programming loop for programming the right half cell. This segmentation allows each loop to operate independently with simplified address inputs, reducing overall programming time while maintaining proper threshold voltage distribution through sequential execution of both loops.
2Ease of manufacture
If conventional single programming loop is used for both half cells, then data can be programmed into multi-level cells, but address input complexity increases
Solution Approach 1:
The patent divides the programming operation into two separate programming loops, where each loop handles a specific half cell with simplified address inputs. The first programming loop programs the left half cell with addresses pointing only to left half cells, and the second programming loop programs the right half cell with addresses pointing only to right half cells. This segmentation eliminates the need for complex address mappings required in conventional single-loop approaches.
Solution Approach 2:
The patent dynamically switches between two different programming modes represented by two separate programming loops. The control logic determines when to execute the first loop versus the second loop based on programming requirements, allowing flexible adaptation to different data sizes and cell configurations while maintaining simple address inputs for each loop.
Data Source
AI summary
A method for programming a memory, which includes multiple multi-level cells each having a left half cell and a right half cell, includes the following steps. First, a target address corresponding to 2n-group data to be stored is provided, wherein n is a positive integer. Next, the 2n-group data is sequentially programmed into the multi-level cells based upon the target address in a programming loop so that the data stored in the left half cells and the data stored in the right half cells are from different groups of the 2n-group data.