Memory Read Correction Using Adjacent Cell Soft Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face challenges in accurately reading data from nonvolatile memory devices due to interference between memory cells, leading to errors in data retrieval.
Innovation Solution
A memory system and method that involves a memory controller to acquire data from selected cells and adjacent cells, applying correction data through soft bit decoding to mitigate interference effects, thereby improving data reading accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data is read directly from memory cells without correction, then reading speed is fast, but data accuracy deteriorates due to inter-cell interference
Solution Approach 1:
The patent applies preliminary action by reading data from adjacent cells before reading the target cell, and using this adjacent cell data to correct the target cell data. The correction process prepares compensation values based on interference patterns from adjacent cells, which are then applied to improve the accuracy of the target cell reading without requiring complex real-time interference cancellation during the main read operation.
2Reliability
If soft bit decoding processing is applied to correct inter-cell interference, then data accuracy is improved, but processing time increases
Solution Approach 1:
The patent segments the data reading process into multiple stages: first reading adjacent cells, then performing soft bit decoding processing on the target cell with interference compensation, and finally combining results. This segmentation allows the complex soft bit decoding to be applied only where needed (on target cells requiring correction) rather than uniformly to all cells, reducing overall processing time while maintaining reliability.
3Measurement precision
If correction data from adjacent cells is used, then interference effects are mitigated, but the number of read operations increases
Solution Approach 1:
The patent merges the reading operations of adjacent cells and target cells into a coordinated sequence, where adjacent cell readings are performed first and their results are immediately utilized to correct target cell readings. This merging approach allows the system to efficiently use the already-read adjacent cell data for correction purposes, avoiding redundant read operations and maintaining reading efficiency while improving precision.
Data Source
AI summary
According to one embodiment, a system includes: a memory device including word lines and memory cells connected to word lines; and a memory controller, the memory controller acquires first data related to threshold voltages of selected cells connected to a selected word line among the word lines, acquires second data from first adjacent cells connected to a first adjacent word line adjacent to one end of the selected word line, acquires first correction data by correcting the first data based on the second data, acquires third data from second adjacent cells connected to a second adjacent word line adjacent to another end of the selected word line, acquires second correction data by correcting the first correction data based on the third data, and executes soft bit decoding processing on the second correction data to generate read data from the selected cells.