News
Ensuring Cross-Device Portability of Electromagnetic Side-Channel Analysis for Digital Forensics
Congratulations to Lojenaa Navanesan and co-authors Nhien-An Le-Khac, Mark Scanlon, Kasun De Zoysa, and Asanka P. Sayakkara on the publication of Ensuring Cross-Device Portability of Electromagnetic Side-Channel Analysis for Digital Forensics in Forensic Science International: Digital Investigation.
Co-authors: Nhien-An Le-Khac, Mark Scanlon, Kasun De Zoysa, and Asanka P. Sayakkara.
AI-generated summary of the contribution: This study addresses the challenge of cross-device portability in Electromagnetic Side-Channel Analysis (EM-SCA) for digital forensics, a crucial aspect of smart device investigation. The authors experimentally investigate the behavior of Machine Learning (ML) models used in EM-SCA across devices from the same make and model, using a wide set of smart devices. The results show that direct application of pre-trained ML models does not yield optimal outcomes, but applying transfer learning techniques leads to high accuracy. This study makes significant contributions to the field of digital forensics, enabling the use of pre-trained models across identical or similar devices.