Share Email Print

Proceedings Paper

Accurate lithography hotspot detection based on PCA-SVM classifier with hierarchical data clustering
Author(s): Jhih-Rong Gao; Bei Yu; David Z. Pan
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

As technology nodes continues shrinking, layout patterns become more sensitive to lithography processes, resulting in lithography hotspots that need to be identified and eliminated during physical verification. In this paper, we propose an accurate hotspot detection approach based on PCA (principle component analysis)-SVM (sup- port vector machine) classifier. Several techniques, including hierarchical data clustering, data balancing, and multi-level training, are provided to enhance performance of the proposed approach. Our approach is accurate and more efficient than conventional time-consuming lithography simulation; in the meanwhile, provides high flexibility to adapt to new lithography processes and rules.

Paper Details

Date Published: 28 March 2014
PDF: 10 pages
Proc. SPIE 9053, Design-Process-Technology Co-optimization for Manufacturability VIII, 90530E (28 March 2014); doi: 10.1117/12.2045888
Show Author Affiliations
Jhih-Rong Gao, Univ. of Texas at Austin (United States)
Bei Yu, Univ. of Texas at Austin (United States)
David Z. Pan, Univ. of Texas at Austin (United States)

Published in SPIE Proceedings Vol. 9053:
Design-Process-Technology Co-optimization for Manufacturability VIII
John L. Sturtevant; Luigi Capodieci, Editor(s)

© SPIE. Terms of Use
Back to Top