.png){kind=icon}
.png){kind=icon}
Return
LeakDiT: Diffusion Transformers for Trace-Augmented Side-Channel Analysis
.png){kind=icon}
PRE
AI
DOI：10.1109/LCA.2025.3639372.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Deep learning has been extensively used in side-channel analysis (SCA), making trace data insufficiency and class imbalance a critical challenge. Although several studies have explored trace augmentation with generative models, two core limitations remain: (i) insufficient integration of SCA domain knowledge into the models and (ii) limited adoption of state-of-the-art diffusion transformers (DiT). This letter presents LeakDiT, a domain-specific one-dimensional DiT that generates high-quality traces. LeakDiT introduces a loss based on normalized inter-class variance (NICV) to produce realistic traces that preserve the leakage structure. Experimental results demonstrate that LeakDiT improves SCA performance and reduces the number of required traces for key recovery.
Keywords:
Training
Transformers
Germanium
Diffusion models
S Box
Electromagnetics
Schedules
Oscilloscopes
Noise reduction
Jitter
Side-channel analysis
data augmentation
deep learning
diffusion transformers
Journal
I
IF:
1.4
Papers:
42
Citations:
781