/split-attack

Deep Learning Analysis for Split Manufactured Layouts with Routing Perturbation

Primary LanguagePythonOtherNOASSERTION

SplitAttack

Split manufacturing of integrated circuits means to delegate the front-end-of-line (FEOL) and back-end-of-line (BEOL) parts to different foundries, in order to prevent overproduction, intellectual property (IP) piracy, or targeted insertion of hardware Trojans. SplitAttack challenges the security promise of split manufacturing by a sophisticated deep neural network that can infer the missing BEOL connections with high accuracy. In paticular, it features following method for an efficient and effective connection prediction:

  • SplitExtract which formulates various layout-level placement and routing hints,
  • a neural network makes use of vector-based and image-based layout features simultaneously,
  • a loss function that directly and effectively select the most probable BEOL connection among the relevant candidates without suffering from an imbalance between positive and negative samples,
  • ...

More details are in the following papers:

1. How to Download

$ git clone https://github.com/cuhk-eda/split-attack

Dependencies

2. How to Run

Toy Test