Multi-phased Flash Programming with Balanced Gray Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for precise programming of multi-level flash memory cells, such as QLC, require additional hardware resources, increasing complexity and cost while reducing available space for other uses due to the need for write buffers to handle foggy-fine programming approaches.
Innovation Solution
Implementing a multi-phased programming method with balanced Gray coding, where each phase programs and extends states to maintain a balanced Gray code, allowing for readable data after the first phase and reducing the need for write buffers by mapping each program state to at least two non-adjacent states in the subsequent phase.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If foggy-fine programming approach is used for precise programming of multi-level cells, then programming precision is improved, but device complexity increases due to additional hardware resources (write buffers)
Solution Approach 1:
The patent extracts and eliminates the write buffer component from the programming system. By using a multi-phased programming approach with balanced Gray coding, the system can achieve precise programming without requiring the additional hardware resources (write buffers) that the traditional foggy-fine approach demands, thus resolving the contradiction between programming precision and device complexity
Solution Approach 2:
The patent changes the programming parameters by implementing a multi-phased programming scheme with balanced Gray coding instead of the traditional foggy-fine approach. This parameter change allows the system to maintain programming precision while reducing hardware requirements, as the balanced Gray code enables direct programming without needing to store intermediate foggy states in write buffers
2Manufacturing precision
If write buffers are added for foggy-fine programming, then programming precision is improved, but available space for other productive uses is reduced
Solution Approach 1:
The patent removes the write buffer requirement from the programming architecture. By employing multi-phased programming with balanced Gray coding, the system achieves precise programming directly in the memory array without needing dedicated write buffer space, thereby increasing the area available for user data and overprovisioning while maintaining programming precision
3Manufacturing precision
If additional hardware resources are allocated for precise programming, then programming precision is improved, but cost increases
Solution Approach 1:
The patent eliminates the need for additional hardware resources (write buffers) by using multi-phased programming with balanced Gray coding. This approach achieves precise programming using only the existing memory array and control logic, thereby reducing manufacturing cost while maintaining programming precision
Solution Approach 2:
The memory array itself serves the dual purpose of storing both the programming data and the final programmed states. The balanced Gray coding scheme enables the memory cells to be programmed directly to their final states through multiple phases without requiring separate write buffer resources, making the system self-sufficient and reducing overall cost
Data Source
AI summary
Disclosed are systems and methods for providing multi-phased programming with balanced Gray coding. A method includes programming, in a first phase, a first portion of data into memory cells of a flash memory in a first-level cell mode. The method also includes retaining, in a cache, at least a subset of the data. The method also includes regenerating the data from at least the cache, wherein the regenerated data includes a second portion of the data. The method also includes programming, in a second phase, the regenerated data in a second-level cell mode based on a mapping from the first-level cell mode to the second-level cell mode. The mapping maps each state distribution in the first-level cell mode to at least two non-adjacent state distributions in the second-level cell mode, and a width of each state distribution in the first-level cell mode may be narrowed.