Notscared:

Notscared is a side-channel analysis framework

⭐ About:

Motivations:

• Study the pros & cons of using Zarr storage for side-channel analysis
• Compare various batching techniques
• Compare various Zarr compression techniques
• Create a flexible & highly parameterized framework for research usage
• Study the effects of multiprocessing on tiled datasets

🟠 Build Status:

Notscared is under development and is not yet ready for public release

In Progress tasks:

• Adjusting zarr file format
  
• Benchmarking compression algorithms
  
• Cleaning up and using pip to create a user friendly experience
  

✨ Code Style:

Conforms to Pep8 Python standards.

⚙️ Underlying Framework:

Utilizes NumPy to compute statistics and Zarr for file storage and manipulation.

✅ Features:

SNR & CPA computation with adjustable:

• processing batch size
• zarr backend
• compression methods
• data selection

✏ How to:

1. Installation: pip install .

2. In your terminal, cd into the top level directory (where this README file is)

Convert .db file to .zarr:
Note: To change what file is converted, change the filename variable in parameters

    python benchmarking/convert_file.py <file_name>

SNR (Signal to Noise Ratio) Demo:
Using tiled example?

    python benchmarking/snr_verification.py <file_name> True

Nontiled?

    python benchmarking/snr_verification.py <file_name> False

CPA (Correlation Power Analysis) Demo:

    python benchmarking/cpa_verification.py <file_name>

Contributors:

• Vincent Immler, Project Creator and Advisor
  
• Aaron Frost, Team Member
  
• Jonah Bosland, Team Member
  
• Matthew Meyer, Team Member
  
• Utkarsh Saxena, Team Member