Building a Framework to Protect Your Privacy from Drones
By Lori Cameron
Published 07/02/2018
Share this on:
Those lost at sea or caught in hostage situations would welcome a team of drones to find them.
However, if those persons were sunbathing in their backyard—perhaps not.
Because of the social and legal consequences of such intrusions, developers now put attention on privacy-protecting systems for unmanned aerial vehicles (UAVs).
Vienna University of Economics and Business researchers propose a privacy framework for UAVs that safeguards restricted areas—such as private property—and prevents the collection of personal data from, say, bystanders in a crowd who happen to be near a police response.
“Drones can cause privacy harms as they can potentially invade people’s private space, and accidentally expose them by processing personal data against their will. Additionally, privacy violations can occur through the unsuspecting collection of information concerning random citizens without any purpose, simply due to constant video recording while flying,” say the authors of “Privacy-Aware Restricted Areas for Unmanned Aerial Systems” in IEEE Security & Privacy. (Login may be required for full text.)
The proposed privacy framework, depicted below, distinguishes between four types of actors: system operators, service providers, citizens, and authentication service providers.
Protecting your privacy from drones: A software system architecture. Four actors (first column) interact with six modules (second and third columns) to provide input to the UAV (fourth column).
For example, a system operator could be a police officer at a control center while the service provider would be the police department. Additionally, the citizen would be anyone wanting to protect his or her privacy.
“Any citizen holding a legal property title may use the system to set their privacy preferences,” the authors say.
The authentication provider is assumed to be a trusted e-identity provider.
These four actors interact with the six different modules in the privacy framework, providing data input to the UAV via the unmanned aerial system (UAS) control program.
The property coordinates must first be represented using a specific geospatial projection and associated with certain attributes, such as specific permissions for flying over a property.
Here’s how it works.
“To enter data, citizens need to identify themselves via an authentication infrastructure, which is offered by an (external) authentication provider. After authentication, the citizens can enter details about their private properties using a web interface that is offered by the service provider. Based on the data input, a checking entity is required to confirm the correctness of the request,” the authors say.
“After the correctness check, convex hulls can be calculated to increase the efficiency of the calculation of the flight path. The system operator can select and request the flight path calculation from the system. If there is no intersection between flight path and restricted areas, the flight path can be submitted to a UAS control program. Finally, the UAS control program handles the communication to the UAV, allowing it to be dispatched according to the flight coordinates chosen,” they add.
The authors of the research are Peter Blank, a process and data analytics professional at PwC Switzerland and a former research assistant at Vienna University of Economics and Business; Sabrina Kirrane, a postdoctoral researcher at the same Vienna University; and Sarah Spiekermann, a professor for business informatics also at the university.
Research related to drones in the Computer Society Digital Library
Lori Cameron is a Senior Writer for the IEEE Computer Society and currently writes regular features for Computer magazine, Computing Edge, and the Computing Now and Magazine Roundup websites. Contact her at l.cameron@computer.org. Follow her on LinkedIn.
