Adaptive Margin Programming for Nonvolatile Memory Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing nonvolatile memory devices face challenges in efficiently programming data due to the absence of certain logic states, leading to reduced reliability and increased read errors, particularly when memory cells have more program states than required by the data, resulting in inefficient margin utilization.
Innovation Solution
A method is introduced where the margins between program states are adjusted based on the absence of certain logic states, with larger margins assigned to more frequently occurring states, and supplemental data is stored to indicate these adjustments, allowing for improved programming and read operations by varying verify and read voltages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If memory cells are configured with multiple program states to increase storage capacity, then storage density is improved, but reliability deteriorates due to increased read errors when certain logic states are absent
Solution Approach 1:
The patent applies dynamics by making the margin distribution adaptive rather than static. The system dynamically adjusts the margins between program states based on the actual logic states present in the data being programmed. When certain logic states are absent, the margins are automatically redistributed to provide larger separation for the states that are actually used, thereby maintaining high reliability while preserving multi-bit storage capacity.
Solution Approach 2:
The patent changes the parameter of margin distribution between program states. Instead of using fixed equal margins, the system varies the margin sizes based on the frequency and presence of different logic states in the program data. This parameter change allows the system to optimize reliability for the specific data pattern being programmed, reducing read errors for absent or rare logic states.
2Ease of manufacture
If equal margins are used between all program states, then manufacturing simplicity is maintained, but reliability deteriorates when certain logic states are absent from program data
Solution Approach 1:
The system transitions from static equal margins to dynamic adaptive margins. The margin configuration is determined at programming time based on analysis of the program data's logic state distribution. This dynamic approach maintains manufacturing simplicity because the same physical memory structure is used, but the logical margin distribution adapts to the data being programmed.
Solution Approach 2:
The patent applies preliminary action by analyzing the program data before programming and determining the optimal margin distribution in advance. The system counts the occurrences of each logic state in the program data, identifies which states are absent or rare, and pre-calculates the adjusted margins before actually programming the memory cells. This preliminary analysis ensures reliability is optimized for the specific data pattern.
3Reliability
If adjusted margins are applied based on logic state frequency, then reliability is improved, but device complexity increases due to additional margin calculation and control logic
Solution Approach 1:
The patent reduces control logic complexity by performing the complex margin calculation work in advance, during the programming phase. The controller analyzes the program data, counts logic state frequencies, and determines the optimal margin distribution before programming begins. Once the margins are determined, the actual programming and subsequent read operations use these pre-calculated margins without requiring complex real-time calculations, thereby maintaining reliability while reducing operational complexity.
Solution Approach 2:
The system applies self-service by having the program data itself provide the information needed to determine the optimal margin distribution. The logic states present in the program data automatically indicate which margins need to be adjusted - absent or rare logic states naturally lead to larger margins for their neighboring states. The system uses the data's own characteristics to guide the margin adjustment, reducing the need for external control complexity.
4Productivity
If all available program states are used, then storage efficiency is maximized, but reliability deteriorates due to insufficient margin separation for absent logic states
Solution Approach 1:
The patent changes the margin parameter dynamically based on the actual usage of program states. When storage efficiency is maximized by using all available program states, the system still maintains reliability by adjusting the margin sizes - giving larger margins to states that are actually present and used, and smaller margins to states that are absent. This selective margin allocation maintains both high storage efficiency and high reliability.
Solution Approach 2:
The patent applies local quality by making margin sizes non-uniform across different program states. Instead of applying a single global margin value, the system assigns different margin sizes to different logic states based on their local characteristics - specifically, whether they are present or absent in the program data. This local differentiation ensures that each state has the appropriate margin for its actual usage, maintaining both efficiency and reliability.
Data Source
AI summary
A method of programming a nonvolatile memory device comprises receiving program data, detecting logic states of the received program data, identifying adjusted margins to be applied to programmed memory cells based on the absence of one or more logic states in the detected logic states, and programming the program data in selected memory cells using the adjusted margins.