PHYSICAL FEATURE MAP- AND DCNN-BASED COMPUTATION METHOD FOR MACHINE LEARNING-BASED INVERSE LITHOGRAPHY TECHNOLOGY SOLUTION

Title:

PHYSICAL FEATURE MAP- AND DCNN-BASED COMPUTATION METHOD FOR MACHINE LEARNING-BASED INVERSE LITHOGRAPHY TECHNOLOGY SOLUTION

Document Type and Number:

WIPO Patent Application WO/2022/016802

Kind Code:

A1

Abstract:

A physical feature map- and DCNN-based computation method for a machine learning-based inverse lithography technology solution, comprising: obtaining N1 feature maps for {Si} (i=1, 2, …, N1) from a lithography target pattern on the basis of an optical ruler Ki(x,y); establishing a neural network model, the neural network model comprising an input layer, a down-sampling layer, a convolution layer, a batch standardization layer, an up-sampling layer, and an output layer, and the input of the neural network model being the feature maps for {Si} (i=1, 2, …, N1); generating an optimal mask image by using a strict inverse lithography technology algorithm to serve as a training target image of neural network training; during neural network model training, performing training and test by means of different combinations of the number of channel dimensions of the input layer, the number N2 of the convolution layers and batch normalization layers, and the number N3 of convolution kernels of each convolution layer and batch normalization layer, until a neural network model of a satisfying combination is found; and inputting N1 feature maps for {Si} (i=1, 2, …, N1) of a wafer lithography target pattern into the trained neural network model to obtain a predicted inverse lithography technology solution pattern.

Inventors:

SHI XUELONG (CN)
YAN YAN (CN)
ZHOU TAO (CN)
YU XUERU (CN)
LI CHEN (CN)

Application Number:

PCT/CN2020/138332

Publication Date:

January 27, 2022

Filing Date:

December 22, 2020

Export Citation:

Click for automatic bibliography generation Help

Assignee:

SHANGHAI IC R & D CENTER CO LTD (CN)
SHANGHAI INTEGRATED CIRCUIT EQUIPMENT & MAT INDUSTRY INNOVATION CENTER CO (CN)

International Classes:

G03F1/36; G06N3/04

Foreign References:

CN111985611A	2020-11-24
CN111310407A	2020-06-19
CN108665060A	2018-10-16
CN107908071A	2018-04-13
CN109976087A	2019-07-05
US20180095359A1	2018-04-05

Attorney, Agent or Firm:

SHANGHAI TIANCHEN INTELLECTUAL PROPERTY AGENCY (CN)

Download PDF:

View/Download PDF PDF Help

Previous Patent: BIOMASS RADIATION-MICROWAVE COUPLED PYROLYSIS SYSTEM AND METHOD

Next Patent: VISUAL POSITIONING METHOD AND APPARATUS, ELECTRONIC DEVICE, AND COMPUTER READABLE STORAGE MEDIUM