What is it about?
We propose a novel method to evaluate abnormal voltage drops in a power supply network without directly measuring the exact voltage values. This method is based on a mechanism that leverages noise to enhance detection accuracy. We demonstrate its effectiveness using clear and comprehensible mathematical reasoning.
Featured Image
Photo by Stephen Dawson on Unsplash
Why is it important?
We propose a novel method to evaluate abnormal voltage drops in power supply networks without the need for direct voltage measurements. This approach introduces a unique mechanism that leverages ambient noise to enhance detection accuracy—challenging the conventional view that noise is always detrimental. What makes our work particularly timely is the growing complexity and miniaturization of modern electronic systems, where direct voltage sensing is increasingly difficult or invasive.
Perspectives
This is my first research paper. To be honest, the discovery of the underlying mechanism was somewhat accidental. We found that datasets containing noise performed significantly better than those without noise. This unexpected result led me to develop the theoretical explanation presented in the paper. It was truly an exciting and enjoyable research experience.
睿超 刘
East China University of Science and Technology
Read the Original
This page is a summary of: Machine Learning-Based Real-Time Detection of Power Analysis Attacks Using Supply Voltage Comparisons, January 2025, ACM (Association for Computing Machinery),
DOI: 10.1145/3658617.3697766.
You can read the full text:
Contributors
The following have contributed to this page
