Over the years, deep learning algorithms have advanced a lot and any innovation in the algorithms are demonstrated and benchmarked for image classification. Several other field including sidechannel analysis (SCA) have recently adopted deep learning with great success. In SCA, the deep learning algorithms are typically working with 1-dimensional (1-D) data. In this work, we propose a unique method to improve deep learning based side-channel analysis by converting the measurements from raw-trace of 1-dimension data based on float or byte data into picture-formatted trace that has information based on the data position. We demonstrate why "Picturization" is more suitable for deep learning and compare how input and hidden layers interact for each raw (1-D) and picture form. As one potential application, we use a Binarized Neural Network (BNN) learning method that relies on a BNN's natural properties to improve variables. In addition, we propose a novel criterion for attack success or failure based on statistical confidence level rather than determination of a correct key using a ranking system.
INDEX TERMSBinarized neural network, Deep learning, Multi-layer perceptron, Non-profiled sidechannel attack
Volatile memory like SDRAM, forms an integral part of any computer system. It stores variety of data including sensitive data like passwords and PIN. The data stored in SDRAM is wiped off on power-off. However, by bringing the RAM to freezing cold temperature before power off, the data can persist for several seconds, allowing recovery through cold boot attacks. In this work, we investigate the vulnerability of IoT device such as Raspberry Pi against cold boot attack for the first time. Our study found that even though the boot sequence is different from laptop, personal computer, and smartphone, we demonstrate that it is still possible to steal the RAM data, even when the bootloader is not public. The net cost of the attack was under 10 dollars and 99.99% of the RAM data was successfully recovered.
Most cryptographic devices should inevitably have a resistance against the threat of side channel attacks. For this, masking and hiding schemes have been proposed since 1999. The security validation of these countermeasures is an ongoing reserach topic, as a wider range of new and existing attack techniques are tested against these countermeasures. This paper examines the side channel security of the balanced encoding countermeasure, whose aim is to process the secret key-related data under a constant Hamming weight and/or Hamming distance leakage. Unlike previous works, we assume that the leakage model coefficients conform to a normal distribution, producing a model with closer fidelity to real-world implementations. We perform analysis on the balanced encoded PRINCE block cipher with simulated leakage model and also an implementation on an AVR board. We consider both standard correlation power analysis (CPA) and bit-wise CPA. We confirm the resistance of the countermeasure against standard CPA, however, we find with a bit-wise CPA that we can reveal the key with only a few thousands traces.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.