SCAUL: Power Side-Channel Analysis with Unsupervised Learning

Ramezanpour, Keyvan; Ampadu, Paul; Diehl, William

doi:10.48550/arxiv.2001.05951

Cited by 2 publications

(2 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, machine learning represents a large area of growth in modern SCA, and has become the primary focus for modern profiling techniques. As machine learning is beyond the scope of this work, we would direct the interested reader to the following works: [57][58][59][60][61][62][63][64][65][66][67][68][69].…”

Section: Practicementioning

confidence: 99%

Power Side-Channel Attack Analysis: A Review of 20 Years of Study for the Layman

Randolph

Diehl

2020

Cryptography

View full text Add to dashboard Cite

Physical cryptographic implementations are vulnerable to so-called side-channel attacks, in which sensitive information can be recovered by analyzing physical phenomena of a device during operation. In this survey, we trace the development of power side-channel analysis of cryptographic implementations over the last twenty years. We provide a foundation by exploring, in depth, several concepts, such as Simple Power Analysis (SPA), Differential Power Analysis (DPA), Template Attacks (TA), Correlation Power Analysis (CPA), Mutual Information Analysis (MIA), and Test Vector Leakage Assessment (TVLA), as well as the theories that underpin them. Our introduction, review, presentation, and survey of topics are provided for the “non expert”, and are ideal for new researchers entering this field. We conclude the work with a brief introduction to the use of test statistics (specifically Welch’s t-test and Pearson’s chi-squared test) as a measure of confidence that a device is leaking secrets through a side-channel and issue a challenge for further exploration.

show abstract

Section: Practicementioning

confidence: 99%

Power Side-Channel Attack Analysis: A Review of 20 Years of Study for the Layman

Randolph

Diehl

2020

Cryptography

View full text Add to dashboard Cite

show abstract

“…An alternative approach is using unsupervised learning to extract the leakage model from the measurements. An SCA attack based on unsupervised learning and sensitivity analysis is introduced in [28]. The leakage model, either in the form a prior knowledge or a training set, constitutes side information required in existing SCA techniques to retrieve information about the secret data.…”

Section: Introductionmentioning

confidence: 99%

SCARL: Side-Channel Analysis with Reinforcement Learning on the Ascon Authenticated Cipher

Ramezanpour,

Ampadu,

Diehl

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Existing side-channel analysis techniques require a leakage model, in the form of a prior knowledge or a set of training data, to establish a relationship between the secret data and the measurements. We introduce side-channel analysis with reinforcement learning (SCARL) capable of extracting data-dependent features of the measurements in an unsupervised learning approach without requiring a prior knowledge on the leakage model. SCARL consists of an auto-encoder to encode the information of power measurements into an internal representation, and a reinforcement learning algorithm to extract information about the secret data. We employ a reinforcement learning algorithm with actor-critic networks, to identify the proper leakage model that results in maximum inter-cluster separation of the auto-encoder representation. SCARL assumes that the lower order components of a generic non-linear leakage model have larger contribution to the leakage of sensitive data. On a lightweight implementation of the Ascon authenticated cipher on the Artix-7 FPGA, SCARL is able to recover the secret key using 24K power traces during the key insertion, or Initialization Stage, of the cipher. We also demonstrate that classical techniques such as DPA and CPA fail to identify the correct key using traditional linear leakage models and more than 40K power traces.CCS Concepts: • Security and privacy → Security in hardware; Hardware attacks and countermeasures; Side-channel analysis and countermeasures; Cryptanalysis and other attacks.

show abstract

SCAUL: Power Side-Channel Analysis with Unsupervised Learning

Cited by 2 publications

References 39 publications

Power Side-Channel Attack Analysis: A Review of 20 Years of Study for the Layman

Power Side-Channel Attack Analysis: A Review of 20 Years of Study for the Layman

SCARL: Side-Channel Analysis with Reinforcement Learning on the Ascon Authenticated Cipher

Contact Info

Product

Resources

About