Attacking Split Manufacturing from a Deep Learning Perspective

TL;DR

This paper demonstrates that deep learning techniques can effectively compromise the security of split manufacturing in integrated circuits by accurately inferring missing connections, challenging its effectiveness against IP theft and hardware Trojans.

Contribution

It introduces a novel deep neural network approach that outperforms existing network-flow attacks in inferring BEOL connections in split manufacturing.

Findings

Achieves 1.21X higher accuracy on M1 splits

Achieves 1.12X higher accuracy on M3 splits

Operates with less than 1% additional runtime

Abstract

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 image-based 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 1.21X accuracy when splitting on M1 and 1.12X accuracy when splitting on M3 with less than 1% running time.