Semiconductor Metrology Data Routing for Real-Time AI Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Semiconductor manufacturing processes face communication bottlenecks that limit the effectiveness of AI-driven metrology, requiring efficient data processing and analysis to provide in-line feedback for real-time adjustments in complex fabrication stages.
Innovation Solution
Integration of a data processing unit (DPU) with metrology equipment, capable of performing AI-based analysis using a GPU or other processors, to differentiate and process data types, sending type A data to the cloud while analyzing type B data locally for immediate feedback and subsequent cloud transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all metrology data is sent to external systems for analysis, then comprehensive analysis can be performed, but communication bottlenecks increase and real-time feedback is delayed
Solution Approach 1:
The patent segments metrology data into two categories: type A data (defect clusters, outliers, electrical power pin images) that is sent directly to external systems, and type B data (raw inspection data requiring intensive processing) that is analyzed locally by the DPU. This segmentation resolves the contradiction by enabling time-sensitive data to be processed locally for real-time feedback while comprehensive data is transmitted for detailed analysis.
Solution Approach 2:
The DPU acts as an intermediary between the metrology equipment and external systems. It performs preliminary analysis of type B data locally and only transmits essential results or selected data to external systems, thereby reducing communication bottlenecks while maintaining analysis comprehensiveness.
2Reliability
If extensive data is transmitted to external systems for AI analysis, then accurate defect detection is achieved, but communication bandwidth is consumed and processing speed is reduced
Solution Approach 1:
The DPU performs preliminary analysis of type B data locally before transmission to external systems. This preliminary processing filters and prepares data in advance, ensuring that only essential information is transmitted, thereby maintaining defect detection accuracy while reducing communication overhead and improving overall processing speed.
3Manufacturing precision
If AI analysis is performed externally for all data types, then comprehensive defect inspection is achieved, but communication overhead increases and real-time adjustments are limited
Solution Approach 1:
The system segments data processing responsibilities between local DPU and external systems. Type A data is handled externally for comprehensive analysis, while type B data is processed locally for immediate feedback. This segmentation enables real-time process adjustments without increasing overall system communication complexity.
Data Source
AI summary
A metrology system includes metrology equipment, a remote communication link, a local communication link, and a data processing unit (DPU). The metrology equipment is configured to generate a stream of data relating to inspected wafers, and to format the generated data into first and second data types. The remote communication link is configured to communicate with an external system. The data processing unit (DPU) is configured to (i) using the remote communication link, send the data belonging to the first data type directly to the external system, and (ii) perform analysis on the data belonging to the second data type, and, using the local communication link, provide results of the analysis to the metrology equipment.