Title: Signal Processing for Time Series Analysis of Software Code Structures
Committee:
Dr. Zajic, Advisor
Dr. Prvulovic, Co-Advisor
Dr. Bloch, Chair
Dr. Anderson
Abstract: The objective of the proposed research is to explore and develop time series analysis techniques for software analysis applications using electromagnetic side-channel signals. In this framework, our first goal is to develop a technique to detect multiple nested periodicities in quasi-periodic signals where we also address the problem of behavioral changes in time. Then, we aim to provide an approach for the detection of dissimilar regions among multiple signals via pattern detection. And, our final objective is to propose a new method to find the similarities through an alignment of multiple signals.