Machine Learning for Hardware Security: Opportunities and Risks

Elnaggar, Rana; Chakrabarty, Krishnendu

doi:10.1007/s10836-018-5726-9

Cited by 64 publications

(20 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We identify the potential HT types at each stage of SoC development and examine the state-of-the-art HT countermeasures related to each stage. The recent success of machine learning (ML) techniques in many research domains has inspired both academic and industrial communities to explore the potential of applying ML to address various HT attacks [11], [56]. A number of achievements utilizing ML for HT defenses have emerged in the last decade.…”

Section: B Motivation and Contributionsmentioning

confidence: 99%

“…However, this important progress has not been systematically reviewed in previous surveys [4], [13], [14], [16], [21], [35], [39]. Elnaggar et al conducted a survey on the applications of ML in the general area of hardware security, including HT attacks, reverse engineering, IC counterfeiting, side-channel attacks, and IC overbuilding [56]. However, a dedicated survey on the applications of ML to HT defenses is still not available in the literature.…”

Section: B Motivation and Contributionsmentioning

confidence: 99%

“…14, it can be concluded that the most extensive use case of ML is HT detection, accounting for approximately 58.3% of all cases. This is because HT detection is the simplest and most common method of identifying HTs, and ML can be inherently and effectively incorporated to address HTs [16], [56]. Furthermore, DFS and bus security have also seen remarkable applications of ML, with percentages of 20.8% and 16.7%, respectively.…”

Section: A Research Trend Analysismentioning

confidence: 99%

“…Ni et al employed the wavelet transform technique to reduce the highfrequency noise and improve the HT detection accuracy from 92.2% to 99.2% [67]. It is suggested in [56] that noisesensitive models such as K-means should be avoided for data training. In addition, model parameter tuning strategies can also be explored to reduce the effects of noise [65].…”

Section: ) Effect Of Process Variations and Noisementioning

confidence: 99%

See 3 more Smart Citations

A Survey on Machine Learning Against Hardware Trojan Attacks: Recent Advances and Challenges

et al. 2020

View full text Add to dashboard Cite

The remarkable success of machine learning (ML) in a variety of research domains has inspired academic and industrial communities to explore its potential to address hardware Trojan (HT) attacks. While numerous works have been published over the past decade, few survey papers, to the best of our knowledge, have systematically reviewed the achievements and analyzed the remaining challenges in this area. To fill this gap, this article surveys ML-based approaches against HT attacks available in the literature. In particular, we first provide a classification of all possible HT attacks and then review recent developments from four perspectives, i.e., HT detection, design-for-security (DFS), bus security, and secure architecture. Based on the review, we further discuss the lessons learned in and challenges arising from previous studies. Despite current work focusing more on chip-layer HT problems, it is notable that novel HT threats are constantly emerging and have evolved beyond chips and to the component, device, and even behavior layers, therein compromising the security and trustworthiness of the overall hardware ecosystem. Therefore, we divide the HT threats into four layers and propose a hardware Trojan defense (HTD) reference model from the perspective of the overall hardware ecosystem, therein categorizing the security threats and requirements in each layer to provide a guideline for future research in this direction.

show abstract

Section: B Motivation and Contributionsmentioning

confidence: 99%

Section: B Motivation and Contributionsmentioning

confidence: 99%

Section: A Research Trend Analysismentioning

confidence: 99%

Section: ) Effect Of Process Variations and Noisementioning

confidence: 99%

See 2 more Smart Citations

A Survey on Machine Learning Against Hardware Trojan Attacks: Recent Advances and Challenges

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Among all existing detection approaches, golden chip-free Trojan detection approaches have attracted wide attention in [2]. Recently, various supervised learning techniques are applied to the golden chip-free Trojan detection approaches [3]. However, the real case is that little knowledge of the Trojan under detection can be obtained, thus, the side-channel traces required in the training sets do not exist in practice.…”

mentioning

confidence: 99%