{"id":4465,"date":"2026-04-09T12:25:00","date_gmt":"2026-04-09T16:25:00","guid":{"rendered":"https:\/\/sites.miamioh.edu\/cse-senior-design-expo\/?p=4465"},"modified":"2026-04-13T15:04:25","modified_gmt":"2026-04-13T19:04:25","slug":"ai-based-side-channel-analysis-for-hardware-security","status":"publish","type":"post","link":"https:\/\/sites.miamioh.edu\/cse-senior-design-expo\/2026\/04\/ai-based-side-channel-analysis-for-hardware-security\/","title":{"rendered":"AI Based Side Channel Analysis for Hardware Security"},"content":{"rendered":"\n
\n\n \n \n \n \n \n \n
Vision<\/td>\n This project exists to address the growing threat of side-channel attacks, where physical leakages such as power usage, electromagnetic emissions, and timing variations are exploited to compromise cryptographic systems. These attacks bypass mathematically secure algorithms by targeting the hardware itself, putting embedded systems and devices at significant risk. Our vision is to advance the field of hardware security by developing AI-driven methods that detect and analyze these vulnerabilities, setting a new standard in guarding critical systems against side-channel threats. <\/td>\n <\/tr>\n\n
Mission<\/td>\n Our mission is to transform a large, externally provided dataset of side-channel traces into a foundation for AI-driven security analysis. The dataset will be carefully cleaned, preprocessed, and structured to support deep learning experimentation. We will apply and train models like BERT, T5, Transformers, and encoder-decoders. By leveraging this dataset, our models will aim to detect patterns of leakage. While specific models may shift, our constant goal is to turn raw side-channel data into reliable, interpretable insights that strengthen hardware security.<\/td>\n <\/tr>\n\n
Team Members<\/td>\n Alex Aussawalaithong & Himank Juttiga<\/td>\n <\/tr>\n \n
Client<\/td>\n Dr. Xianglong Feng<\/td>\n <\/tr>\n\n
Site<\/td>\n N\/A<\/span><\/td>\n <\/tr>\n <\/table>\n\n