Electro-Magnetic Side-Channel Attack Through Learned Denoising and Classification
This addresses security vulnerabilities in information systems where data must be kept secret, offering a practical attack and defense solution, though it appears incremental by combining existing techniques like deep learning with side-channel analysis.
The paper tackles the problem of reconstructing intercepted data from electro-magnetic side-channel signals by introducing a system that uses deep learning and character recognition to automatically retrieve over 57% of characters in real-time, regardless of signal type, and extends it to a protection system with over 95% success rate in detecting compromises.
This paper proposes an upgraded electro-magnetic side-channel attack that automatically reconstructs the intercepted data. A novel system is introduced, running in parallel with leakage signal interception and catching compromising data in real-time. Based on deep learning and character recognition the proposed system retrieves more than 57% of characters present in intercepted signals regardless of signal type: analog or digital. The approach is also extended to a protection system that triggers an alarm if the system is compromised, demonstrating a success rate over 95%. Based on software-defined radio and graphics processing unit architectures, this solution can be easily deployed onto existing information systems where information shall be kept secret.