ClouDroid : cloud-assisted privacy-preserved data sharing in mobile social networks

There has been an emerging development of location-based social networks on mobile phones, that allow users in dispersed locations to share information such as places, photos and captions from the area, and generally communicate while on the move. The nature of data shared on this platform is highly dynamic, changing rapidly with user coordinates and time of day as a user moves from place to place. Given the sensitive nature of data such as their location, a user would expect a comprehensive set of privacy options that allow them to carefully control who has access to their information. A cloudbased provider could assist in such data sharing, because of its scalability that guarantees the storage of large volumes of data, and its high computational power which allows for the dynamic calculation and configuration of privacy constraints. We study a set of scenarios that represent the most common data sharing needs on a location-based network, and identify extensions that must be made to basic access control mechanisms to enable these scenarios. We then develop a cloud-based data provider that can gather location data and privacy settings from various mobile clients and dynamically generate access control policies for sharing with the rest of the world. In developing this provider, we extend the popular XACML access-control framework, which is able to handle most basic usage needs, and then must be extended accordingly to fit our identified scenarios. The cloud provider is accessed through a basic mobile client that we develop in order to showcase common use cases.We subsequently evaluate the performance of the cloud-based provider in typical contexts.Bachelor of Engineering (Computer Science