Configuration and Collection Factors for Side-Channel Disassembly

Myriad uses, methodologies, and channels have been explored for side-channel analysis. However, specific implementation considerations are often unpublished. This paper explores select test configuration and collection parameters, such as input voltage, shunt resistance, sample rate, and microcontroller clock frequency, along with their impact on side-channel analysis performance. The analysis use case considered is instruction disassembly and classification using the microcontroller power side-channel. An ATmega328P microcontroller and a subset of the AVR instruction set are used in the experiments as the Device Under Test (DUT). A time-series convolutional neural network (CNN) is used to evaluate classification performance at clock-cycle fidelity. We conclude that configuration and collection parameters have a meaningful impact on performance, especially where the instruction-trace's signal to noise ratio (SNR) is impacted. Additionally, data collection and analysis well above the Nyquist rate is required for side-channel disassembly. We also found that 7V input voltage with 1 kiloohm shunt and a sample rate of 250-500 MSa/s provided optimal performance in our application, with diminishing returns or in some cases degradation at higher levels.

Wavelet Selection and Employment for Side-Channel Disassembly

Side-channel analysis, originally used in cryptanalysis is growing in use cases, both offensive and defensive. Wavelet analysis is a commonly employed time-frequency analysis technique used across disciplines, with a variety of purposes, and has shown increasing prevalence within side-channel literature. This paper explores wavelet selection and analysis parameters for use in side-channel analysis, particularly power side-channel-based instruction disassembly and classification. Experiments are conducted on an ATmega328P microcontroller and a subset of the AVR instruction set. Classification performance is evaluated with a time-series convolutional neural network (CNN) at clock-cycle fidelity. This work demonstrates that wavelet selection and employment parameters have meaningful impact on analysis outcomes. Practitioners should make informed decisions and consider optimizing these factors similarly to machine learning architecture and hyperparameters. We conclude that the gaus1 wavelet with scales 1-21 and grayscale colormap provided the best balance of classification performance, time, and memory efficiency in our application.