Dataset for: Hound: Locating Cryptographic Primitives in Desynchronized Side-Channel Traces Using Deep-Learning

This dataset is part of "Hound: Locating Cryptographic Primitives in Desynchronized Side-Channel Traces Using Deep-Learning" [1] available online.

The source code for testing the dataset is available on GitHub.

The dataset is organized as follows:

/training: Contains three subsets: train, valid, and test. Each subset consists of two .npy files:

_set: Contains the preprocessed side-channel traces.

_labels: Contains the target labels for training the CNN, labeling each data as `CP start`, `CP spare`, or `noise`.

/inference: Contains files for two demos: consecutive AES executions and AES executions interleaved with noisy applications. Each demo consists of two .npy files:

aes_: Contains the side-channel traces to input into Hound.

gt_: Contains the ground truth for checking the correctness of Hound segmentation.

This repository is protected by copyright and licensed under the Creative Commons Attribution 4.0 International license.

© 2024 hardware-fab

[1] D. Galli, G. Chiari and D. Zoni, "Hound: Locating Cryptographic Primitives in Desynchronized Side-Channel Traces using Deep-Learning," 2024 IEEE 42nd International Conference on Computer Design (ICCD), Milan, Italy, 2024, pp. 114-121, doi: 10.1109/ICCD63220.2024.00027.