Intelligent Machine Homicide

Heuser, Annelie; Zohner, Michael

doi:10.1007/978-3-642-29912-4_18

Cited by 115 publications

(44 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In all cases template attack either matched or outperformed LS-SVM. Better results were achieved in Heuser and Zohner (2012), where the HW was classified successfully with SVM, requiring fewer attack traces with moderate to high noise.…”

Section: Introductionmentioning

confidence: 88%

Random Forest profiling attack on advanced encryption standard

Patel

Baldwin

2014

IJACT

View full text Add to dashboard Cite

Random Forest, a non-parametric classifier, is proposed for byte-wise profiling attack on advanced encryption standard (AES) and shown to improve results on PIC microcontrollers, especially in high-dimensional variable spaces. It is shown in this research that data collected from 40 PIC microcontrollers exhibited highly non-Gaussian variables. For the full-dimensional dataset consisting of 50,000 variables, Random Forest correctly extracted all 16 bytes of the AES key. For a reduced set of 2,700 variables captured during the first round of the encryption, Random Forest achieved success rates as high as 100% for cross-device attacks on 40 PIC microcontrollers from four different device families. With further dimensionality reduction, Random Forest still outperformed classical template attack for this dataset, requiring fewer traces and achieving higher success rates with lower misclassification rate. The importance of analysing the system noise in choosing a classifier for profiling attack is examined and demonstrated through this work. His area of specialty is computer security, computer communication networks, and wireless network protocols with an emphasis on real-time analysis. He has published over 90 articles in the area of security, wireless networks, wireless protocols, and queuing theory as well as numerous conference papers in the area. He has been awarded two patents and four USAF inventions for his work in computer security. His current research interests include computer communication networks, computer security, cyber operations and reconfigurable computing systems.

show abstract

Section: Introductionmentioning

confidence: 88%

Random Forest profiling attack on advanced encryption standard

Patel

Baldwin

2014

IJACT

View full text Add to dashboard Cite

show abstract

“…Recent work [8]- [10] investigated machine learning techniques, and in particular Support Vector Machines (SVM), as an alternative to classical template building approaches. SVM are probably the most prominent member of kernel methods.…”

Section: A Previous Workmentioning

confidence: 99%

“…While classical templates aim at building an explicit characterisation of data, SVM aim at separating data into classes, but clearly noise will negatively impact the performance of both techniques; other intuitive parameters are the number of available traces, and the number of relevant points as described above. When the noise is higher, SVM outperform classical approaches, and require slightly fewer traces to succeed [10].…”

Section: A Previous Workmentioning

confidence: 99%

Reliable Information Extraction for Single Trace Attacks

Banciu

Oswald

Whitnall

2015

Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE), 2015

View full text Add to dashboard Cite

Abstract-Side-channel attacks using only a single trace crucially rely on the capability of reliably extracting side-channel information (e.g. Hamming weights of intermediate target values) from traces. In particular, in original versions of simple power analysis (SPA) or algebraic side channel attacks (ASCA) it was assumed that an adversary can correctly extract the Hamming weight values for all the intermediates used in an attack. Recent developments in error tolerant SPA style attacks relax this unrealistic requirement on the information extraction and bring renewed interest to the topic of template building or training suitable machine learning classifiers.In this work we ask which classifiers or methods, if any, are most likely to return the true Hamming weight among their first (say s) ranked outputs. We experiment on two data sets with different leakage characteristics. Our experiments show that the most suitable classifiers to reach the required performance for pragmatic SPA attacks are Gaussian templates, Support Vector Machines and Random Forests, across the two data sets that we considered. We found no configuration that was able to satisfy the requirements of an error tolerant ASCA in case of complex leakage.

show abstract

“…They first compared TA and learning algorithms, namely Random Forest (RF), SVM, and Self-Organizing Maps (SOM), and then proposed an enhanced brute force algorithm to break the key. Heuser et al [25] analyzed multiple bits of the key based on the Hamming weight model by using multi-class SVM. The authors divided the intermediate power consumption into several classes and then calculated the probability of belonging to each class.…”

Section: Introductionmentioning

confidence: 99%

“…The previous work [23][24][25][26][27] focused more on how SVM translates a problem of breaking the key into the classification of machine learning. There is no systematic literature to study the elements that influence the performance of SVM in power analysis attacks.…”

Section: Introductionmentioning

confidence: 99%