TUM Logo

Machine Learning in Side-Channel Analysis

Machine Learning in Side-Channel Analysis

Supervisor(s): Emanuele Strieder
Status: open
Topic: Machine Learning Methods
Type of Thesis: Masterthesis Bachelorthesis Guided Research
Thesis topic in co-operation with the Fraunhofer Institute for Applied and Integrated Security AISEC, Garching


Master’s thesis, Research Assistant, Internship, Bachelor’s thesis
Machine Learning in Side-Channel Analysis
Utilizing statistical techniques, side-channel analysis exploits information that a cryptographic device is leaking. Possible sources of this leakage are electromagnetic or power side-channel traces.
Machine learning based side-channel analysis extends the statistical toolbox with Neural Networks, Belief Propagation or different methods of this field to recombine and exploit leakage.
In collaboration with the Technical University of Munich, the Fraunhofer AISEC’s hardware security department offers a variety of open positions in this field. Depending on your strengths, we provide both pure software-based and practical hardware topics, such as the following:
• Trace analysis using explainable machine learning
• Leakage recombination using belief propagation - light-weight or post-quantum algorithms
• Belief propagation performance optimization using GPUs
• Pattern-based triggering using software-defined radios

On request, other topics can be offered.

• Programming skills, at least one language (Python, C, Rust)
• Interest in hardware security
• Basic Linux skills

Emanuele Strieder
Telefon: +49 89 322-9986-140
E-Mail: emanuele.strieder@aisec.fraunhofer.de
Fraunhofer Research Institution for Applied and Integrated Security (AISEC)
Department Hardware Security
Parkring 4, 85748 Garching (near Munich), Germany