Memory Cell Soft Data Combination for Retrieval Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory technologies face challenges in accurately determining the quality and confidence of data stored in multi-level memory cells, particularly when cells store fractional bits, as individual cell soft data may not provide sufficient information for evaluating the reliability of stored data.
Innovation Solution
The solution involves an array of memory cells, where each cell can be programmed to multiple program states, and a controller determines soft data associated with these states to evaluate the quality and confidence of the data stored, using a combination of soft data from multiple cells to enhance accuracy and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If individual cell soft data is used to evaluate data quality, then the process is simple, but the accuracy and reliability of data evaluation is insufficient
Solution Approach 1:
The patent combines soft data from multiple memory cells to evaluate the quality and confidence of stored data. Instead of relying on individual cell soft data, the system merges information from several cells, thereby improving measurement precision while accepting increased processing complexity as a necessary trade-off for reliable data evaluation in multi-level memory cells.
2Quantity of substance
If multi-level memory cells store fractional bits, then memory density increases, but the reliability of individual cell soft data decreases
Solution Approach 1:
The patent addresses the reduced reliability of individual cell soft data in multi-level memory cells by combining soft data from multiple cells. This merging approach compensates for the uncertainty introduced by fractional bit storage, maintaining data reliability while achieving high memory density through multi-level cell architecture.
Solution Approach 2:
The system uses combined soft data from multiple cells as an intermediary to evaluate the reliability of stored information. Rather than directly trusting individual cell soft data, the patent introduces a combination process that acts as a mediator to filter out uncertainties and provide more reliable data quality assessment.
3Measurement precision
If soft data from multiple cells is combined, then accuracy and reliability of data retrieval increases, but processing complexity increases
Solution Approach 1:
The patent implements a systematic approach to merging soft data from multiple memory cells, combining their information to improve data retrieval accuracy. The method accepts increased processing complexity as a necessary investment to achieve more precise and reliable data evaluation, particularly for multi-level memory cells where individual cell data is insufficient.
Data Source
AI summary
The present disclosure includes apparatuses and methods for determining soft data for combinations of memory cells. A number of embodiments include an array of memory cells including a first and second memory cell each programmable to one of a number of program states, wherein a combination of the program states of the first and second memory cells corresponds to one of a number of data states, and a buffer and/or a controller coupled to the array and configured to determine soft data associated with the program states of the first and second memory cells and soft data associated with the data state that corresponds to the combination of the program states of the first and second memory cells based, at least in part, on the soft data associated with the program states of the first and second memory cells.