Injected and Leaked: Actively Inducing Side-channel Leakage Using Electromagnetic Injection And Hardware Non-Linearity
Abstract
This deposit contains the Open Science artifacts for InjectEave. It includes case-study audio data and a speech denoising/enhancement codebase to reproduce the reported experiments and demonstrate performance under different recording environments. Contents - Case Study1&2: 6 clean original utterances (3 male + 3 female) and their received recordings captured in two environments (home and meeting room; 6 files each, .wav). - Case Study3: 5 original sound 5 speaker reference utterances, 5 received recordings, and 5 denoised outputs (all .wav). - Denoise Code: SGMSE-based speech denoising training and inference code, with configs and preprocessing scripts (see Denoise Code/README.md).
Fields of Study & Subjects
Cryptographic Implementations and Security
Speech and Audio Processing
Security and Verification in Computing
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 5 February 2026