METHOD AND SYSTEM FOR SYSTEMATIC DEFECT IDENTIFICATION

• US 20120170830A1
• Filed: 12/29/2010
• Published: 07/05/2012
• Est. Priority Date: 06/14/2010
• Status: Active Grant

• Current Assignee: Carnegie Mellon University
• Original Assignee: Carnegie Mellon University
• Inventors: Blanton, Ronald DeShawn, Tam, Wing Chiu
• Granted Patent: US 8,509,517 B2
• Time in Patent Office: Days
• US Class Current: 382/149
• CPC Class Codes: G06F 18/23213 with fixed number of clusters, e.g. K-means clustering
• CPC Class Codes: G06F 30/398 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 Class Codes: G06T 2207/30148 Semiconductor; IC; Wafer
• CPC Class Codes: G06T 7/001 using an image reference approach

Abstract

A method and apparatus for identifying suspect layout features from a plurality of layout features of an integrated circuit (IC) layout. A plurality of snippet images is generated, each of which depicts at least a portion of a suspect layout feature which is different from suspect layout features depicted in others of the plurality of snippet images. The suspect layout features are determined based on the diagnosis of a plurality of defective ICs manufactured in accordance with the IC layout. A plurality of clusters is generated, and each of the clusters contains a group of the plurality of snippet images based on similarities between the snippet images.

Claims

1. A method for identifying a suspect layout feature from a plurality of layout features contained in an integrated circuit (IC) layout, comprising:

generating a plurality of snippet images depicting suspect polygons, each of the snippet images depicting at least a portion of a suspect polygon which is different from a portion of the suspect polygon depicted in others of the plurality of snippet images, the suspect polygons being determined based on the diagnosis of a plurality of defective ICs manufactured according to the IC layout; and

assigning each of the plurality of snippet images to a particular cluster of a plurality of clusters based on similarities between snippet images in each respective cluster.

2. The method of claim 1, further comprising:

identifying a plurality of suspect nets from the plurality of defective ICs, wherein each of the suspect polygons is contained in at least one of the plurality of suspect nets.

3. The method of claim 2, wherein assigning each of the plurality of snippet images to the particular cluster of the plurality of clusters based on the similarities between the snippet images in the each respective cluster further comprises:

in a first pass, for each suspect net of the plurality of suspect nets, assigning each of the plurality of snippet images that depicts at least a portion of a suspect polygon from the each suspect net to one of a plurality of first clusters;

selecting a plurality of representative snippet images from ones of the plurality of first clusters; and

in a second pass, assigning each of the plurality of representative snippet images to the particular cluster of the plurality of clusters based on similarities between the representative snippet images in each respective cluster.

4. The method of claim 3, wherein selecting the representative snippet images from the ones of the plurality of first clusters comprises selecting the representative snippet images from the ones of the plurality of first clusters such that no two snippet images from a same first cluster are selected as representative snippet images.

5. The method of claim 2, wherein assigning each of the plurality of snippet images to the particular cluster of the plurality of clusters based on the similarities between the snippet images in the each respective cluster further comprises:

in a first pass, for each defective IC of the plurality of defective ICs, assigning each of a plurality of snippet images generated based on a diagnosis of the each defective IC to one of a plurality of first clusters;

selecting a plurality of representative snippet images from ones of the plurality of first clusters; and

in a second pass, assigning each of the plurality of representative snippet images to the particular cluster of the plurality of clusters based on similarities between the representative snippet images in the each respective cluster.

6. A method for identifying suspect layout features from a plurality of layout features of an integrated circuit (IC) layout, comprising:

generating a plurality of snippet images, each of the snippet images depicting at least a portion of a suspect layout feature which is different from suspect layout features depicted in others of the plurality of snippet images, the suspect layout features being determined based on the diagnosis of a plurality of defective ICs manufactured in accordance with the IC layout; and

generating a plurality of clusters, each of the clusters containing a group of the plurality of snippet images based on similarities between the snippet images.

7. The method of claim 6, further comprising:

identifying a primary cluster from the plurality of clusters having a largest group of the plurality of snippet images; and

identifying a subset of snippet images from the primary cluster closest to a center of the primary cluster.

8. The method of claim 7, further comprising:

performing a hotspot analysis on the subset of snippet images to determine if any of the snippet images in the subset of snippet images contains a hotspot.

9. The method of claim 8, wherein the hotspot analysis is performed by a lithography simulation module.

10. The method of claim 6, wherein each of the suspect layout features comprises at least a portion of a suspect polygon.

11. The method of claim 10, wherein the portion of the suspect polygon depicted in each of the plurality of snippet images is positioned in the respective snippet image with respect to a same reference location.

12. The method of claim 11, wherein the same reference location is a center of the respective snippet image.

13. The method of claim 6, wherein each of the plurality of snippet images has the same dimensions and the same scale.

14. The method of claim 6, wherein at least some of the plurality of snippet images further depict a second layout feature in addition to the at least the portion of the suspect layout feature.

15. A computing device comprising:

a communications interface adapted to communicate with a network; and

a control system comprising a processor and coupled to the communications interface, the control system adapted to;

generate a plurality of snippet images, each of the snippet images depicting at least a portion of a suspect layout feature which is different from suspect layout features depicted in others of the plurality of snippet images, the suspect layout features being determined based on diagnoses of a plurality of defective integrated circuits manufactured in accordance with an integrated circuit (IC) layout; and

generate a plurality of clusters, each of the clusters containing a group of the plurality of snippet images based on similarities between the plurality of snippet images.

16. The computing device of claim 15, wherein the control system is further adapted to:

identify a primary cluster from the plurality of clusters having a largest group of the plurality of snippet images; and

identify a subset of snippet images from the primary cluster closest to a center of the primary cluster.

17. The computing device of claim 16, wherein the control system is further adapted to:

perform a hotspot analysis on the subset of snippet images to determine if any of the snippet images in the subset of snippet images contains a hotspot.

18. The computing device of claim 15, wherein each of the suspect layout features comprises at least a portion of a suspect polygon.

19. The computing device of claim 18, wherein the portion of the suspect polygon depicted in each of the plurality of snippet images is positioned in the respective snippet image with respect to a same reference location.

20. The computing device of claim 15, wherein at least some of the plurality of snippet images further depict a second layout feature in addition to the at least the portion of the suspect layout feature.