Security Vulnerabilities of Unmanned Aerial Vehicles and Countermeasures: An Experimental Study

Vishal Dey, Vikramkumar Pudi, Anupam Chattopadhyay, Yuval Elovici

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

25 Scopus citations

Abstract

Drones present a novel airborne platform for new commercial and consumer tasks. Due to their increasing use, security analysis has become necessary. In this paper, we present a collection of fatal attacks, minor observations and associated security vulnerabilities for DJI Phantom 4 Pro(P4P) and Parrot Bebop 2 drones. We also propose some countermeasures to bolster the security of the analyzed drones. In addition to this, we provided a brief overview of the system and embedded architecture of these vehicles. Although DJI has tried to improve security by introducing radio communication and partially encrypted firmware, the P4P still remains vulnerable to GPS spoofing, jamming and other vulnerabilities through DJI-SDK has been exposed in our experimental study. In contrast, Bebop drones are more vulnerable to wireless attacks and we have experimentally validated this for Bebop 2 drone.

Original languageEnglish
Title of host publicationProceedings - 31st International Conference on VLSI Design, VLSID 2018 - Held concurrently with 17th International Conference on Embedded Systems, ES 2018
PublisherIEEE Computer Society
Pages398-403
Number of pages6
ISBN (Electronic)9781538636923
DOIs
StatePublished - 27 Mar 2018
Event31st International Conference on VLSI Design, VLSID 2018 - Pune, India
Duration: 6 Jan 201810 Jan 2018

Publication series

NameProceedings of the IEEE International Conference on VLSI Design
Volume2018-January
ISSN (Print)1063-9667

Conference

Conference31st International Conference on VLSI Design, VLSID 2018
Country/TerritoryIndia
CityPune
Period6/01/1810/01/18

Fingerprint

Dive into the research topics of 'Security Vulnerabilities of Unmanned Aerial Vehicles and Countermeasures: An Experimental Study'. Together they form a unique fingerprint.

Cite this