SkyDummy: Personal Drone-Assisted Dummy Generation Strategies for Preserving Location Privacy

Low-altitude, small-sized, and user-friendly drones are rapidly emerging and have been deployed in diverse civilian and military applications. We envision that each user will be able to carry a personal drone and conduct a drone-based operation. Unlike prior literature, defending against a drone-centric privacy attack and misbehavior, we newly leverage a drone to protect against privacy leakage in location-based services (LBS) querying. In this paper, we propose drone-assisted dummy generation strategies, Distance and Lazy, to protect the location privacy of the user from the server. The basic idea is that both the user and drone collaborate to generate and select a set of fake querying locations to confuse the server. The drone flies around the user by following one of the mobility models, Distance Close, Controlled Random, and Circular, and supports the user's anonymity. The proposed strategies increase the level of anonymity by extending the user's cloaking area. The Lazy approach also judiciously generates dummies in strategic positions and extends the size of the cloaking area. We conduct extensive simulation experiments for performance evaluation and comparison in terms of the size of the combined cloaking area and the ratio of duplicated cloaking area. Our extensive evaluations show that the proposed strategies can improve the location privacy of the users by extending and shaping the anonymous area and be a viable privacy-preserving mechanism in the upcoming personal drone era.