Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A neural network model and optical proximity correction technology, applied in the field of optical proximity correction based on neural network model, can solve the problems of increased computing time, impractical full-chip implementation, etc., and achieve the effect of computing improvement
Active Publication Date: 2018-04-13
SHANGHAI INTEGRATED CIRCUIT RES & DEV CENT
View PDF6 Cites 19 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
All inverse lithography methods have a large increase in computation time, therefore full-chip implementations of inverse lithography solutions remain impractical
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0057] When the neural network model is a linear neural network model, the required photolithographic mask pattern, that is, the auxiliary pattern and the corrected pattern of the main pattern, can be regarded as at a certain threshold, by cutting a continuous class intensity function and the obtained contour. This class of intensity functions can be derived from the optical image intensity function I(x,y) of the lithographic target pattern via a fixed nonlinear mapping mechanism. Obviously, It not only depends on I(x, y), but also depends on the gray distribution of the optical image intensity function I(x, y) around the (x, y) point. The most efficient way to encode the gray level distribution of the optical image intensity function I(x,y) around the point (x,y) is to use a set of values of the intrinsic imaging signal at the point (x,y). In order to accurately describe the optical image intensity including 3D effects of lithographic masks, the intrinsic imaging signal...
Embodiment 2
[0083] When the neural network model is a quadratic neural network model, as attached Figure 4As shown, the lithographic mask plane is divided into small units, assuming that t(i, j) and t(m, n) are behind the small unit (i, j) and small unit (m, n) lithographic mask light field. Since the response of chemical photoresists is the light intensity, not the light field itself, we can imagine that the target pattern calculated from the reverse lithography technique can be cut by a continuous intensity-like function of and the obtained contour. . This function only depends on all pair values {t(i,j),t(m,n)}, but the function itself is unknown:
[0084]
[0085] Pairs {t(i,j),t(m,n)} are defined around the point (x,y). {t(i,j),t(m,n)} are real quantities because all lithography mask types in use today have only phase 0 or phase 180. One way to explore this unknown function is to use a multi-layer perceptron neural network model, with the set {t(1,1)*t(1,1),t(1,1)*t(1,2),...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention discloses a neural network model-based optical proximity correction method. The neural network model-based optical proximity correction method comprises the following steps of S01, training a neural network model, wherein M test patterns are selected on a training optical mask, target patterns corresponding to the M test patterns are respectively obtained, a similar strength functionI-bar (X, Y) is simulated by a known perception neural network, and the perception neural network is trained according to the similar strength function and the target patterns to obtain the neutral network model; and S02, achieving optical proximity correction by the trained neutral network model, in which the similar strength function I-bar (X, Y) of an optical mask to be processed is obtained by the obtained neutral network model, the similar strength function I-bar (X, Y) is cut by a cutting threshold to generate the optical mask containing the target patterns, and photoetching is performed by employing the optical mask containing the target patterns as a mask after the optical proximity correction. The optical proximity correction method disclosed by the invention is simultaneously compatible with corrected image quality and relatively rapid implementation speed.
Description
technical field [0001] The invention relates to the field of optical proximity correction, in particular to an optical proximity correction method based on a neural network model. Background technique [0002] Optical proximity correction (OPC) has become an essential means in the semiconductor manufacturing process. Its purpose is to make the pattern realized on the chip as consistent as possible with the target pattern of lithography through photolithography mask pattern correction. OPC is composed of several key steps, such as target pattern setting in lithography, auxiliary pattern generation and main pattern correction. Due to the deviation caused by etching or the requirements of the lithography process window, the target pattern of lithography is often different from the original design pattern. Auxiliary patterns are lithography process windows used to improve sparse design patterns, and their placement rules are often derived from lithography simulations. The mai...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more