TY - GEN
T1 - Attacking split manufacturing from a deep learning perspective
AU - Li, Haocheng
AU - Patnaik, Satwik
AU - Sengupta, Abhrajit
AU - Yang, Haoyu
AU - Knechtel, Johann
AU - Yu, Bei
AU - Young, Evangeline F.Y.
AU - Sinanoglu, Ozgur
N1 - Funding Information:
This work is supported in part by The Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. CUHK14202218) and Center for Cyber Security Abu Dhabi (CCS-AD) in New York University Abu Dhabi.
Publisher Copyright:
© 2019 Association for Computing Machinery.
PY - 2019/6/2
Y1 - 2019/6/2
N2 - The notion of integrated circuit split manufacturing which delegates the front-end-of-line (FEOL) and back-end-of-line (BEOL) parts to different foundries, is to prevent overproduction, piracy of the intellectual property (IP), or targeted insertion of hardware Trojans by adversaries in the FEOL facility. In this work, we challenge the security promise of split manufacturing by formulating various layout-level placement and routing hints as vector- and imagebased features. We construct a sophisticated deep neural network which can infer the missing BEOL connections with high accuracy. Compared with the publicly available network-flow attack [1], for the same set of ISCAS-85 benchmarks, we achieve 1.21× accuracy when splitting on M1 and 1.12× accuracy when splitting on M3 with less than 1% running time.
AB - The notion of integrated circuit split manufacturing which delegates the front-end-of-line (FEOL) and back-end-of-line (BEOL) parts to different foundries, is to prevent overproduction, piracy of the intellectual property (IP), or targeted insertion of hardware Trojans by adversaries in the FEOL facility. In this work, we challenge the security promise of split manufacturing by formulating various layout-level placement and routing hints as vector- and imagebased features. We construct a sophisticated deep neural network which can infer the missing BEOL connections with high accuracy. Compared with the publicly available network-flow attack [1], for the same set of ISCAS-85 benchmarks, we achieve 1.21× accuracy when splitting on M1 and 1.12× accuracy when splitting on M3 with less than 1% running time.
UR - http://www.scopus.com/inward/record.url?scp=85067831688&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85067831688&partnerID=8YFLogxK
U2 - 10.1145/3316781.3317780
DO - 10.1145/3316781.3317780
M3 - Conference contribution
AN - SCOPUS:85067831688
T3 - Proceedings - Design Automation Conference
BT - Proceedings of the 56th Annual Design Automation Conference 2019, DAC 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 56th Annual Design Automation Conference, DAC 2019
Y2 - 2 June 2019 through 6 June 2019
ER -