Memory Error Correction Using Erasure-Marked Cell Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices face challenges in error detection and correction, particularly in high-density memory systems where read disturb and shorts can lead to erroneous data sensing, reducing performance and lifetime, and conventional error correction methods may not be sufficient to handle increased error rates without altering algorithms or reducing data volume.
Innovation Solution
The implementation of a method that identifies and marks memory cells unable to store data as erasures, using techniques such as secondary sense operations, bipolar sense operations, and sensing circuitry with check signals to enhance error correction capabilities by distinguishing between valid and invalid data states, thereby improving the correction rate of error correction operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional error correction methods are used in high-density memory systems, then memory density is increased, but error rates increase due to read disturb and shorts
Solution Approach 1:
The patent applies preliminary action by performing a first sense operation before the main read operation to identify memory cells that cannot store data. This advance detection allows the system to mark these cells as erasures before attempting error correction, improving the overall error correction capability without requiring changes to the memory density or architecture.
Solution Approach 2:
The patent introduces an intermediary mechanism by using check signals and sense operations that act as mediators between the memory cells and the error correction process. These intermediary sense operations detect problematic cells and provide information to the error correction algorithm, enabling more effective correction without directly altering the memory density or error characteristics.
2Productivity
If memory cells that cannot store data are not identified, then sense operation speed is maintained, but error correction capability is reduced
Solution Approach 1:
The patent applies partial action by performing sense operations on only certain memory cells (those suspected of being unable to store data) rather than all cells. This selective sensing approach maintains overall sense operation speed while providing sufficient information to improve error correction capability for the problematic cells.
Solution Approach 2:
The patent uses preliminary sense operations to identify problematic memory cells before the main read operation. This advance identification allows the error correction algorithm to focus resources on correcting errors in identified cells, improving correction capability without significantly impacting overall sense operation speed.
3Reliability
If secondary sense operations are performed to identify memory cells unable to store data, then error correction rate is improved, but sense operation time is increased
Solution Approach 1:
The patent applies partial action by performing secondary sense operations only on memory cells that are suspected of being unable to store data, rather than on all memory cells. This selective approach improves error correction rate by focusing detection efforts where needed, while minimizing the additional time required for sense operations.
Solution Approach 2:
The patent applies skipping by rapidly performing sense operations on identified problematic cells and then proceeding with error correction. The sense operations are executed efficiently to minimize time loss, and once problematic cells are identified, the process moves quickly through the correction phase without unnecessary delays.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach increases the error correction capability by accurately identifying and correcting errors, particularly in high-density memory systems, by marking erasures and using these markings to recover original data, thus enhancing the raw bit error rate and overall memory performance.
Implementation Method 1
A state of a resistance variable memory cell can be determined by sensing current through the cell responsive to an applied interrogation voltage. The sensed current, which varies based on the resistance level of the cell, can indicate the state of the cell.
Data Source
AI summary
The present disclosure includes apparatuses, methods, and systems for error detection and correction in memory. An embodiment includes a memory having a group of self-selecting memory cells which store data corresponding to a codeword from an error correcting code, and circuitry configured to perform a sense operation on the group of self-selecting memory cells, identify, based on the sense operation, memory cells of the group that cannot store data, mark data sensed from the identified memory cells as erasures and perform an error correction operation on data sensed from the group of self-selecting memory cells with the data sensed from the identified memory cells marked as erasures.