Memory Cell Repeating Groups for Automated Layout Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual classification of memory cell groups based on layer stacking structures is time-consuming and prone to high error rates, leading to inaccurate analyses of product yield and failure modes.
Innovation Solution
A method utilizing a pattern recognition apparatus to rapidly and accurately classify memory cells by recognizing their layout patterns, dividing them into groups, and establishing memory cell repeating groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual classification of memory cell groups is performed based on layer stacking structures, then memory cells can be classified into groups, but the process consumes considerable time and is prone to high error rates
Solution Approach 1:
The patent replaces the manual mechanical classification process with an automated pattern recognition apparatus that uses image processing and algorithmic analysis to identify and classify memory cell groups based on layout patterns, thereby eliminating time consumption and human error associated with manual classification
Solution Approach 2:
The patent creates a digital representation (layout pattern image) of the memory cell array structure and processes this copy through pattern recognition algorithms, allowing rapid classification without physically manipulating or manually examining each memory cell, thus significantly reducing classification time while maintaining accuracy
2Reliability
If manual classification of memory cell groups is performed, then memory cell groups can be established, but the process is prone to high error rates leading to inaccurate analyses
Solution Approach 1:
The patent substitutes the error-prone manual classification process with an automated pattern recognition system that consistently applies predefined algorithms to identify memory cell groups, eliminating human error and improving the reliability of subsequent product yield and failure mode analyses
Solution Approach 2:
The pattern recognition apparatus incorporates feedback mechanisms where classification results are validated against expected patterns, and the system can iteratively refine its recognition accuracy, thereby ensuring high reliability in memory cell group classification and subsequent analysis
Data Source
AI summary
A method for establishing a memory cell group including different memory cells includes following steps. A memory cell array including memory cells and a layout pattern of the memory cell array are provided. A pattern recognition apparatus for performing a recognition step on the layout pattern is provided. The recognition step includes: dividing the memory cells in an nth row of rows in the memory cell array into row groups, where the row groups are identical, and the memory cells in the row group are different; determining whether preceding n rows and succeeding n rows of the rows are repeated. According to a recognition result provided by the pattern recognition apparatus through performing the recognition step, a portion of the memory cells is collectively established as a memory cell repeating group, so that the memory cell array is composed of a plurality of the memory cell repeating groups.