Feasibility of Low-cost Drone Direction Finding via Software-defined Radio

Abstract

In this thesis, the feasibility of a low-cost solution for the direction finding of first-person view drones by their radio emissions is explored. The increased widespread use of fast-flying first-person view drones has highlighted the need for affordable detection systems to address the risks they pose, such as physical harm and privacy violations. Existing solutions to this problem are often not affordable. A mechanically-actuated scanner utilising a directional antenna and a software-defined radio device was designed for direction finding of first-person view drone video signals. Evaluation of the designed system at distances ranging from 25m to 250m showed little average error when determining the direction of a first-person view drone's video signal emissions relative to the built scanner. Töö eesmärk oli uurida tarkvaralise raadio abil raallidroonide raadiolokatsiooni taskukohast teostatavust rallidroonide poolt välja saadetud signaalide põhjal. Rallidroonide laialdane levik ja nende poolne oht tekitada kahju nii varale kui ka inimeste tervisele ja privaatsusele, on toonud esile vajaduse tuvastussüsteemide järele. Olemasolevad lahendused ei ole sageli taskukohased. Seetõttu loodi lahendus, mis võimaldas rallidroonide signaali suunda detektori suhtes leida kasutades motoriseeritud suundantenni ja tarkvaralist raadiot. Seadme testimisel leiti, et juhul kui droon oli kaugustel 25m kuni 250m, oli keskmine viga rallidrooni suuna leidmisel detektorist relatiivselt väike.