Detecting hotspots in physical design layout patterns utilizing hotspot detection model with data augmentation

Patent metadata

Field	Value
Publication number	US-10706205-B2
Application number	US-201816166892-A
Country	US
Kind code	B2
Filing date	Oct 22, 2018
Priority date	Oct 22, 2018
Publication date	Jul 7, 2020
Grant date	Jul 7, 2020

Abstract

A method for detecting hotspots in physical design layout patterns includes receiving a given physical design layout pattern, utilizing a hotspot detection model to detect one or more potential hotspots in the given physical design layout pattern, and performing a verification to determine whether a given potential hotspot of the one or more potential hotspots detected by the hotspot detection model comprises a real hotspot or a nonexistent hotspot. The method also includes, responsive to determining that the given potential hotspot comprises an actual hotspot, modifying the given physical design layout pattern to remove the actual hotspot. The method further includes, responsive to determining that the given potential hotspot comprises a nonexistent hotspot, augmenting the hotspot detection model with additional training data generated based on the nonexistent hotspot.

First claim

What is claimed is:

1. A method for detecting hotspots in physical design layout patterns, the method comprising steps of:
receiving a given physical design layout pattern;
utilizing a hotspot detection model to detect one or more potential hotspots in the given physical design layout pattern;
performing a verification to determine whether a given potential hotspot of the one or more potential hotspots detected by the hotspot detection model comprises a real hotspot or a nonexistent hotspot;
responsive to determining that the given potential hotspot comprises an actual hotspot, modifying the given physical design layout pattern to remove the actual hotspot; and
responsive to determining that the given potential hotspot comprises a nonexistent hotspot, augmenting the hotspot detection model with additional training data generated based on the nonexistent hotspot;
wherein augmenting the hotspot detection model with additional training data generated based on the nonexistent hotspot comprises:
modifying one or more features in the given physical design layout pattern to generate one or more additional training layout patterns; and
training the hotspot detection model with the one or more additional training layout patterns; and

wherein the method is performed by at least one processing device comprising a processor coupled to a memory.

2. The method of claim 1 , wherein a hotspot represents an area of a physical design layout pattern that at least one of: is not manufacturable; and renders at least a portion of a patterned structure comprising the hotspot inoperable.

3. The method of claim 2 , wherein the hotspot detection model comprises a machine learning model.

4. The method of claim 3 , wherein the machine learning model comprises a convolutional neural network (CNN) model.

5. The method of claim 1 , wherein the performing a verification comprises performing optical proximity correction (OPC) simulation of the given physical design layout pattern.

6. The method of claim 1 , wherein the performing a verification comprises performing a wafer verification of the given physical design layout pattern.

7. The method of claim 1 , wherein the modifying the one or more features in the given physical design layout pattern comprises adjusting a location of at least one given feature of the one or more features in the given physical design layout pattern until the given physical design layout pattern comprises an actual hotspot.

8. The method of claim 7 , further comprising:
generating at least a first additional training layout pattern comprising the given physical design layout pattern marked as a hotspot-free physical design layout pattern; and
generating at least a second additional training layout pattern comprising the adjusted given physical design layout pattern marked as a hotspot-containing physical design layout pattern.

9. The method of claim 7 , wherein adjusting the given feature comprises iteratively adjusting the location of the given feature in designated increments until the given physical design layout pattern comprises an actual hotspot.

10. The method of claim 9 , wherein the iteratively adjusting the location of the given feature comprises adjusting the location of the given feature in a first dimension and a second dimension of the given physical design layout pattern.

11. The method of claim 10 , further comprising:
generating at least a first additional training layout pattern comprising a negative adjustment of the location of the given feature in the first dimension;
generating at least a second additional training layout pattern comprising a positive adjustment of the location of the given feature in the first dimension;
generating at least a third additional training layout pattern comprising a negative adjustment of the location of the given feature in the second dimension; and
generating at least a fourth additional training layout pattern comprising a positive adjustment of the location of the given feature in the second dimension.

12. The method of claim 11 , wherein one of the first additional training layout pattern and the second additional training layout pattern is marked as hotspot-containing and the other one of the first additional training layout pattern and the second additional training layout pattern is marked as hotspot-free, and wherein one of the third additional training layout pattern and the fourth additional training layout pattern is marked as hotspot-containing and the other one of the third additional training layout pattern and the fourth additional training layout pattern is marked as hotspot-free.

13. The method of claim 7 , wherein the modifying the one or more features in the given physical design layout pattern is automated based on a type of hotspot that the hotspot detection model is configured to detect.

14. The method of claim 13 , wherein the hotspot detection model is configured to detect bridging hotspots, and wherein the modifying the one or more features in the given physical design layout pattern comprises automatically reducing a spacing between a first feature and a second feature in the given physical design layout pattern until a bridging hotspot is detected.

15. The method of claim 13 , wherein the hotspot detection model is configured to detect line end shortening hotspots, and wherein the modifying the one or more features in the given physical design layout comprises narrowing a line end width of a given feature until a line end width hotspot is detected.

16. The method of claim 1 , further comprising evaluating manufacturability of a given patterned structure comprising the given physical design layout pattern.

17. The method of claim 1 , further comprising repeating the modifying and the augmenting until no potential hotspots are detected for the given physical design layout pattern.

18. A computer program product, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by at least one computing device to cause the at least one computing device to perform steps of:
receiving a given physical design layout pattern;
utilizing a hotspot detection model to detect one or more potential hotspots in the given physical design layout pattern;
performing a verification to determine whether a given potential hotspot of the one or more potential hotspots detected by the hotspot detection model comprises a real hotspot or a nonexistent hotspot;
responsive to determining that the given potential hotspot comprises an actual hotspot, modifying the given physical design layout pattern to remove the actual hotspot; and
responsive to determining that the given potential hotspot comprises a nonexistent hotspot, augmenting the hotspot detection model with additional training data generated based on the nonexistent hotspot;
wherein augmenting the hotspot detection model with additional training data generated based on the nonexistent hotspot comprises:
modifying one or more features in the given physical design layout pattern to generate one or more additional training layout patterns; and
training the hotspot detection model with the one or more additional training layout patterns.

19. An apparatus comprising:
a memory; and
at least one processor coupled to the memory and configured for:
receiving a given physical design layout pattern;
utilizing a hotspot detection model to detect one or more potential hotspots in the given physical design lay

Assignees

• IBM

Inventors

• Shao Dongbing
• Sha Jing
• Lai Kafai

Classifications

• G03F7/705 Primary

  Modelling or simulating from physical phenomena up to complete wafer processes or whole workflow in wafer productions · CPC title

• G06F30/398 Primary

  Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM] (optical proximity correction [OPC] design processes G03F1/36) · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06F18/2411

  based on the proximity to a decision surface, e.g. support vector machines · CPC title