Electronic devices may undergo attacks going beyond traditional cryptanalysis. Side-channel analysis (SCA) is an alternative attack that exploits information leaking from physical implementations of e.g. cryptographic devices to discover cryptographic keys or other secrets. This work comprehensively investigates the application of a machine learning technique in SCA. The considered technique is a powerful kernel-based learning algorithm: the Least Squares Support Vector Machine (LS-SVM). The chosen side-channel is the power consumption and the target is a software implementation of the Advanced Encryption Standard. In this study, the LS-SVM technique is compared to Template Attacks. The results show that the choice of parameters of the machine learning technique strongly impacts the performance of the classification. In contrast, the number of power traces and time instants does not influence the results in the same proportion. This effect can be attributed to the usage of data sets with straightforward Hamming weight leakages in this first study.

Implementations of cryptographic primitives are vulnerable to physical attacks. While the adversary only needs to succeed in one out of many attack methods, the designers have to consider all the known attacks, whenever applicable to their system, simultaneously. Thus, keeping an organized, complete and up-to-date table of physical attacks and countermeasures is of paramount importance to system designers.This paper summarizes known physical attacks and countermeasures on Elliptic Curve Cryptosystems. Instead of repeating the details of different attacks, we focus on a systematic way of organizing and understanding known attacks and countermeasures. Three principles of selecting countermeasures to thwart multiple attacks are given. This paper can be used as a road map for countermeasure selection in a first design iteration.

Increased complexity in modern embedded systems has presented various important challenges with regard to side-channel attacks. In particular, it is common to deploy SoC-based target devices with high clock frequencies in security-critical scenarios; understanding how such features align with techniques more often deployed against simpler devices is vital from both destructive (i.e., attack) and constructive (i.e., evaluation and/or countermeasure) perspectives. In this paper, we investigate electromagnetic-based leakage from three different means of executing cryptographic workloads (including the general purpose ARM core, an on-chip co-processor, and the NEON core) on the AM335x SoC. Our conclusion is that addressing challenges of the type above is feasible, and that key recovery attacks can be conducted with modest resources.

No abstract

The design of a reader antenna is described for usage in Radio Frequency IDentification (RFID) systems at 13.56 Mhz, as defined in the ISO-14443a standard. It presents the theory, with emphasis on the effect of the read out distance on the design, but also describes measurements on concrete designs to validate the formulas and statements. We also comment on practical problems that were encountered during the design process. The major contribution of this work is the generalization of the design theory for large read out distances where the conventional assumption of constant loop current no longer holds. Index Terms-HF radio communication, loop antennas, magnetic fields, Q factor. 1 Activation range is defined as the distance from the reader where the field is still large enough to power up the tag. 2 Even if 10 cm is often suggested as the maximum range, the ISO-14443 standard does not mention this.