|Student Name:||Capt Benjamin Kuhar|
|Thesis:||RF Emitter Tracking and Intent Assessment|
|Location:||ENG Conference Room, Building 640 Room 317|
|Date & Time:||02/27/2013 at 1000|
|Abstract:|| Current research in employing pattern recognition techniques in a Wireless Sensor Network (WSN) to detect anomalous or suspicious behavior is limited. The purpose of this research was to determine the feasibility of an accurate tracking and intent assessment system of unknown or foreign radio frequency (RF) emitters in close proximity to and within military installations as a method for physical security. 22 position tracks were collected using a hand-held Global Positioning System (GPS) unit and feature data from five different features were generated for each track. Each collected position track was individually classified in Matrix Laboratory(MATLAB®) using the leave-one-out-cross-validation (LOOCV) method for four different classification methods. The four classification methods used in this research were the linear discriminant analysis (LDA), the diagonal linear discriminant analysis (DLDA),the quadratic discriminant analysis (QDA), and the Mahalanobis method. The accuracies and false positive and false negative error rates of the four classification methods were compared. Additionally, best fit receiver operating characteristic (ROC) curves were generated for each classification method and discussed. In this research, the QDA classification method out-performed the other three classification methods.