Date of Award

3-2022

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Electrical and Computer Engineering

First Advisor

Barry E. Mullins, PhD

Abstract

The growing presence of Unmanned Aerial Vehicle (UAV) brings new threats to the civilian and military front. In response, the Department of Defense (DoD) is developing many drone detection systems. Current systems use Radio Detection and Ranging (RADAR), Light Detection and Ranging (LiDAR), and Radio Frequency (RF). Although useful, these technologies are becoming easier to spoof every year, and some are limited to line of sight. Acoustic emissions are a unique quality all drones emit. Acoustics are difficult to spoof and do not require line of sight for detection. This research expands the research field of study by creating HurtzHunter, a prototype which tests acoustic payload detection at far range (7 m - 100 m) and with cell phone devices. HurtzHunter uses MFCCs to train a SVM for UAV acoustic payload detection. Depending on the recording device and SVM configuration, the results show an 82-98% payload prediction accuracy using cell phone devices.

AFIT Designator

AFIT-ENG-MS-22-M-026

DTIC Accession Number

AD1166859

Share

COinS