An AI Architecture with the Capability to Classify and Explain Hardware Trojans

TL;DR

This paper presents an explainable AI architecture for hardware Trojan detection that not only classifies suspected circuits but also provides explanations for its decisions, enhancing transparency in digital hardware security.

Contribution

It introduces a novel explainable methodology and architecture for hardware Trojan detection based on existing detection features, with results demonstrating its effectiveness on trust-hub benchmarks.

Findings

Effective explanation of hardware Trojans in netlists

Improved transparency in Trojan detection decisions

Validated on trust-hub benchmark circuits

Abstract

Hardware trojan detection methods, based on machine learning (ML) techniques, mainly identify suspected circuits but lack the ability to explain how the decision was arrived at. An explainable methodology and architecture is introduced based on the existing hardware trojan detection features. Results are provided for explaining digital hardware trojans within a netlist using trust-hub trojan benchmarks.