Personal drone-assisted dummy gernation strategies for enhancing location privacy

With ubiquitous high-speed wireless networks and resource-rich smartphones, users can access Internet data and services anywhere and enjoy location-based services (LBS) by querying points of interest (POIs). However, users often trade their location privacy with services by unconsciously sharing the location information with an LBS server without knowing that the server might not be fully trusted. In light of this, we propose drone-assisted dummy generation strategies to protect the location privacy of users from the server. The basic idea is that both the user and drone collaborate to create a set of fake querying locations to confuse the server. When a user generates an LBS query, it launches a drone that flies around the user by following a mobility pattern. In this thesis, we envision that each user will carry a personal drone embedded into a smartphone and conduct a drone-based operation. First, we investigate passive and active drone mobility models and visualize the traces with the user's movements for analysis: (i) distance close, (ii) controlled random, and (iii) circular. Second, we propose a lazy approach in dummy generation to increase the level of anonymity by efficiently increasing the size of cloaking areas, created by both user and drone. We conduct extensive simulation experiments using the Matlab software environment for performance evaluation and analysis in terms of the combined cloaking area size and ratio of duplicated cloaking areas. The simulation results show that the proposed strategies can improve the location privacy of the users by judiciously shaping the cloaking area and be a viable dummy generation approach in the upcoming personal drone era.