Charged Particle Beam Writing Data Processing Pipeline

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing data processing methods for charged particle beam writing systems face challenges in reducing Turn Around Time (TAT) due to the sequential processing of chip and layout data, which leads to inefficiencies in data registration and increased processing time.

Innovation Solution

A data processing method that divides design data into frame-basis pipeline processing for conversion and preprocessing, allowing for the registration of writing data in a charged particle beam writing apparatus on a frame basis, thereby reducing TAT by enabling simultaneous processing and registration of multiple frames.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If writing data is registered on a layout basis after external apparatus completes processing on all chip data and layout data, then data registration can be performed in the writing apparatus, but the Turn Around Time (TAT) of the writing process cannot be reduced

Engineering Contradiction:
ImproveTurn Around Time (TAT)VSAvoiddata registration speed
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The patent segments the data registration process by dividing chip data into multiple frames and registering them individually rather than waiting for complete layout data processing. This allows frame-by-frame registration to proceed in parallel with ongoing data processing, reducing the overall Turn Around Time while maintaining data integrity through proper synchronization.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary registration of chip data frames in the writing apparatus before the external apparatus completes processing of all layout data. By registering frames preliminarily and updating them as processing completes, the system reduces idle time and accelerates the overall writing process without compromising accuracy.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If all chip data and layout data are processed by the external apparatus before writing apparatus starts data registration, then data accuracy is maintained, but processing speed is reduced

Engineering Contradiction:
Improvedata accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The patent maintains continuous useful action by enabling the writing apparatus to register chip data frames as soon as they are processed by the external apparatus, rather than waiting for complete processing. This continuous registration process operates in parallel with ongoing data processing, maintaining high utilization of the writing apparatus while ensuring data accuracy through synchronized updates.

Inventive Principle:
Principle #20Continuity of useful action

Solution Approach 2:

The patent introduces dynamic data registration where the registration process adapts to the processing status of the external apparatus. Frames are registered dynamically as they become available, with the system continuously adjusting the registration workflow to maintain both speed and accuracy without rigid sequential constraints.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS10714312B2Data processing method, charged particle beam writing apparatus, and charged particle beam writing system
Publication Date: 2020.07.14 NUFLARE TECH INC
  • US10714312B2 patent drawing
  • US10714312B2 patent drawing
  • US10714312B2 patent drawing

AI summary

In one embodiment, a data processing method is provided for generating writing data from design data and registering the writing data in a charged particle beam writing apparatus. The method includes generating the writing data by performing a plurality of conversion processes on a plurality of pieces of first frame data obtained through division of the design data corresponding to one chip, and performing a plurality of preprocessing processes on a plurality of pieces of second frame data obtained through division of the writing data of the chip, and registering the writing data of the chip in the charged particle beam writing apparatus. The plurality of conversion processes are performed in frame-basis pipeline processing, and the plurality of preprocessing processes are performed in frame-basis pipeline processing. The writing data is registered in the charged particle beam writing apparatus on a frame basis.