David Petrizze – Thesis Defense

About the event

Student:  David Petrizze

Degree:  M.S. Electrical Engineering

Committee Chair:  Dr. Sandip Roy

Thesis Title:  A Hypothesis Testing Approach to Low-Overhead Trajectory-Based Classification of Aerial Intruders

Abstract:  Motivated by security concerns in the growing use of unmanned aerial vehicles (UAVs), a method for classifying aerial intrusions based on intrinsic flight characteristics is presented. The presented method utilizes a coarse physical model to aid in the classification of measured data; a hypothesis testing problem is posed wherein aerial intruders are modelled as seeking to regulate their speed on a linear trajectory in the presence of a noisy disturbance (e.g. wind), and a maximum a posteriori (MAP) detector is developed to classify intruders as a function of measured velocity samples. A simplified form of the detector is determined, which fundamentally relies on two points in the sample autocorrelation of the input, and readily lends itself toward a linear classifier. Further, methods for computing the probability of error for given intruder models are presented, and the probability error is shown to tend toward zero as the number of samples is increased. Finally, the detector is simulated for synthesized data, and considerations for its practical implementation are briefly discussed.