A first of its kind technique for locating a photographed target illegally using drones was revealed in a study published by information security researchers from Ben-Gurion University of the Negev and the Weizmann Institute. The study focused on the growing concerns of privacy intrusion due to the increased use of drones for personal and business applications.
In the new article, "The drone games - target detection from an encrypted video transmission", The researchers present techniques used to locate a target photographed by a drone while intercepting and analyzing the encrypted traffic.
"The beauty of this research is that by simply connecting to a laptop and controlling a certain flashing lighting device, it is possible to find out if a drone is photographing a specific object (person/house)". Says Ben Nashi, a doctoral student under the guidance of Prof. Yuval Elovitz in the Department of Software Engineering and Information Systems at Ben-Gurion University and a researcher at the Cyber Security Research Center. Prof. Elovitz is the head of the institute, as well Head of the Deutsche Telekom Innovation Labs at Ben-Gurion University. "Whereas in the past it was possible to locate a drone, now it is possible to say whether the use of the drone's camera was done against a specific goal."
In the first simulation, the researchers showed how it is possible to detect an attempted invasion of a person's privacy in his home. For this, they used a smart curtain that they placed on the window of the house and intercepted the encrypted video traffic that is transmitted from the drone to the operator, which is transmitted on the FPV (First person view) channel. By flashing the smart curtain that caused an increase in the transmission rate of the video traffic from the drone, the researchers were able to prove that the neighbor using a DJI Mavic drone was filming their home and was illegally using the drone that violated their privacy.
In another simulation they did, the researchers presented how an LED strip attached to a white shirt could be used to detect illegal drone activity. The researchers demonstrated how the blinking of the LED lights on the shirt can be controlled, in a way that affects the transmission rate of the FPV channel and causes it to modulate the word "SOS" in binary.
"Game Of Drones":
"This research debunks the common belief that using encryption to secure an FPV channel prevents attackers from knowing they are being spied on," says Nashi. "The secret behind the method is to force controlled changes to the photographed target to affect the transmission rate of the data transmitted on the FPV channel."
The method can be used from any laptop running the Linux OS and does not require complex hacking abilities or cryptographic skills.
"Our findings may help in thwarting privacy invasion attacks using drones that are gradually becoming common. Apart from civilian effects, the research also has military effects," explains Nashi. It was recently announced that the IDF also uses drones of the type used by the researchers (DJI MAVIC) in their research.
In a previous study in the field of information security and drones, Naseeh showed how A laser attached to a drone of sorts 1 Can initiate code execution with DJI Inspire malware installed on an isolated network through the scanner.
The research team included Raz Ben Nathanel, a student in the Department of Communication Systems Engineering, and Prof. Adi Shamir (Turing Award winner) from the Weizmann Institute who originally raised the idea in a workshop organized by the information security laboratories of Ben-Gurion University in the field of Internet of Things security.
3 תגובות
You can "bypass" this test easily. If you move the camera from side to side or even zoom in and out alternately to keep the bitrate high. That way this whole method is invalid
When writing a technical article, it is imperative that the writer understands what he is writing.
It's enough that another car was driving in the frame and it no longer works