Far-field light scattering from sub-wavelength wafer patterns using a parallel FETI-DP algorithm

Kedi Zhang, Jinlong Zhu, Lynford L. Goddard, Jian Ming Jin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The dual-primal finite-element and interconnecting algorithm (FETI-DP) is efficiently applied in parallel to compute the far-field scattering from a wafer with sub-wavelength patterns. As a nonoverlapping domain decomposition method for solving large-scale problems, the FETI-DP algorithm formulates a global interface problem, whose iterative solution is accelerated by solving a global corner problem. To maintain good parallel scaling efficiency, global communication is minimized by using an iterative classical Gram-Schmidt scheme for the global interface problem and a communication-avoiding approach for the global corner problem. In addition, a preconditioner for the corner system matrix is implemented to improve iterative convergence. After computing the far-field data, digital postprocessing is used to remove the background signal and improve the detection sensitivity of isolated defects.

Original languageEnglish (US)
Title of host publication2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2009-2010
Number of pages2
ISBN (Electronic)9781509028863
DOIs
StatePublished - Oct 25 2016
Event2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Fajardo, Puerto Rico
Duration: Jun 26 2016Jul 1 2016

Publication series

Name2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings

Other

Other2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016
CountryPuerto Rico
CityFajardo
Period6/26/167/1/16

Fingerprint

Light scattering
far fields
light scattering
wafers
Domain decomposition methods
Wavelength
Communication
communication
wavelengths
iterative solution
digital data
Scattering
Defects
decomposition
scaling
defects
matrices
scattering

ASJC Scopus subject areas

  • Instrumentation
  • Radiation
  • Computer Networks and Communications

Cite this

Zhang, K., Zhu, J., Goddard, L. L., & Jin, J. M. (2016). Far-field light scattering from sub-wavelength wafer patterns using a parallel FETI-DP algorithm. In 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings (pp. 2009-2010). [7696711] (2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/APS.2016.7696711

Far-field light scattering from sub-wavelength wafer patterns using a parallel FETI-DP algorithm. / Zhang, Kedi; Zhu, Jinlong; Goddard, Lynford L.; Jin, Jian Ming.

2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. p. 2009-2010 7696711 (2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhang, K, Zhu, J, Goddard, LL & Jin, JM 2016, Far-field light scattering from sub-wavelength wafer patterns using a parallel FETI-DP algorithm. in 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings., 7696711, 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 2009-2010, 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016, Fajardo, Puerto Rico, 6/26/16. https://doi.org/10.1109/APS.2016.7696711
Zhang K, Zhu J, Goddard LL, Jin JM. Far-field light scattering from sub-wavelength wafer patterns using a parallel FETI-DP algorithm. In 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2016. p. 2009-2010. 7696711. (2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings). https://doi.org/10.1109/APS.2016.7696711
Zhang, Kedi ; Zhu, Jinlong ; Goddard, Lynford L. ; Jin, Jian Ming. / Far-field light scattering from sub-wavelength wafer patterns using a parallel FETI-DP algorithm. 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 2009-2010 (2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings).
@inproceedings{d0b6b1d3b06f4121a31f7db49ad10d84,
title = "Far-field light scattering from sub-wavelength wafer patterns using a parallel FETI-DP algorithm",
abstract = "The dual-primal finite-element and interconnecting algorithm (FETI-DP) is efficiently applied in parallel to compute the far-field scattering from a wafer with sub-wavelength patterns. As a nonoverlapping domain decomposition method for solving large-scale problems, the FETI-DP algorithm formulates a global interface problem, whose iterative solution is accelerated by solving a global corner problem. To maintain good parallel scaling efficiency, global communication is minimized by using an iterative classical Gram-Schmidt scheme for the global interface problem and a communication-avoiding approach for the global corner problem. In addition, a preconditioner for the corner system matrix is implemented to improve iterative convergence. After computing the far-field data, digital postprocessing is used to remove the background signal and improve the detection sensitivity of isolated defects.",
author = "Kedi Zhang and Jinlong Zhu and Goddard, {Lynford L.} and Jin, {Jian Ming}",
year = "2016",
month = "10",
day = "25",
doi = "10.1109/APS.2016.7696711",
language = "English (US)",
series = "2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "2009--2010",
booktitle = "2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings",
address = "United States",

}

TY - GEN

T1 - Far-field light scattering from sub-wavelength wafer patterns using a parallel FETI-DP algorithm

AU - Zhang, Kedi

AU - Zhu, Jinlong

AU - Goddard, Lynford L.

AU - Jin, Jian Ming

PY - 2016/10/25

Y1 - 2016/10/25

N2 - The dual-primal finite-element and interconnecting algorithm (FETI-DP) is efficiently applied in parallel to compute the far-field scattering from a wafer with sub-wavelength patterns. As a nonoverlapping domain decomposition method for solving large-scale problems, the FETI-DP algorithm formulates a global interface problem, whose iterative solution is accelerated by solving a global corner problem. To maintain good parallel scaling efficiency, global communication is minimized by using an iterative classical Gram-Schmidt scheme for the global interface problem and a communication-avoiding approach for the global corner problem. In addition, a preconditioner for the corner system matrix is implemented to improve iterative convergence. After computing the far-field data, digital postprocessing is used to remove the background signal and improve the detection sensitivity of isolated defects.

AB - The dual-primal finite-element and interconnecting algorithm (FETI-DP) is efficiently applied in parallel to compute the far-field scattering from a wafer with sub-wavelength patterns. As a nonoverlapping domain decomposition method for solving large-scale problems, the FETI-DP algorithm formulates a global interface problem, whose iterative solution is accelerated by solving a global corner problem. To maintain good parallel scaling efficiency, global communication is minimized by using an iterative classical Gram-Schmidt scheme for the global interface problem and a communication-avoiding approach for the global corner problem. In addition, a preconditioner for the corner system matrix is implemented to improve iterative convergence. After computing the far-field data, digital postprocessing is used to remove the background signal and improve the detection sensitivity of isolated defects.

UR - http://www.scopus.com/inward/record.url?scp=84997206930&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84997206930&partnerID=8YFLogxK

U2 - 10.1109/APS.2016.7696711

DO - 10.1109/APS.2016.7696711

M3 - Conference contribution

AN - SCOPUS:84997206930

T3 - 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings

SP - 2009

EP - 2010

BT - 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -