Multi-Channel Memory Read Decisions for MLC Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices face challenges in error correction, particularly in Multi-Level Cell (MLC) and Multi-Bit Cell (MBC) memory devices, where error occurrence rates are high, leading to increased hardware complexity and read failure rates due to overlapping threshold voltage distributions and reduced voltage windows.
Innovation Solution
A memory device with a decision unit that performs both hard and soft decisions on data read from multiple channels, using optimized numbers of decision levels based on channel characteristics to enhance error correction probability and reduce hardware complexity, by applying Error Control Codes (ECC) decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ECC encoding/decoding methods are applied to overcome errors in MLC/MBC memory devices, then error correction capability is improved, but hardware complexity increases
Solution Approach 1:
The patent segments the data reading process into multiple channels, where each channel is read using an optimized number of decision levels based on its specific characteristics. This segmentation allows the system to apply ECC decoding selectively to channels with higher error rates, rather than uniformly across all channels, thereby reducing overall hardware complexity while maintaining error correction capability for critical data paths.
Solution Approach 2:
The patent applies different numbers of decision levels to different channels based on their individual characteristics. Channels with higher error probabilities receive more decision levels (soft decision), while channels with lower error probabilities use fewer decision levels (hard decision). This local optimization reduces the total hardware complexity while maintaining reliability where it is most needed.
2Device complexity
If a fixed number of decision levels is used for all channels, then device complexity is reduced, but error correction probability decreases
Solution Approach 1:
The patent introduces dynamic adaptability by setting different numbers of decision levels for different channels based on their characteristics. Instead of a static, uniform approach, the system dynamically adjusts the decision level count per channel, allowing optimal error correction probability for each channel while managing overall device complexity.
Solution Approach 2:
The patent changes the parameter of decision level count from a fixed value to a variable that depends on channel characteristics. By adjusting this parameter based on each channel's error probability and data importance, the system achieves higher error correction probability without uniformly increasing device complexity across all channels.
Data Source
AI summary
Disclosed are a memory device and a data decision method. The memory device may include a memory cell array, and a decision unit configured to read first data from the memory cell array via a first channel, perform at least one of a hard and soft decision on the first data using a first number of decision levels set based on characteristics of the first channel, read second data from the memory cell array via a second channel, and perform a soft decision on the second data using a second number of decision levels set based on characteristics of the second channel.