Lightweight Software Product Line Based Privacy Protection Scheme for Pervasive Applications

Protecting user's privacy is one of the main concerns for the deployment of pervasive computing systems in the real world. In pervasive environments, the user context information is naturally distributed among different spatial or logical domains. Many efforts have been done to match the service privacy policy with the user's privacy preferences. However, since the pervasive environments are characterized by a large number of available services as well as a large amount of context information, the privacy protection mechanism poses two main requirements. Firstly, policies are created on a per task basis. We argue here that specifying the privacy on a per domain basis facilitates specifying the privacy preferences for the user. Secondly, to ease specifying the user' privacy preferences, an intuitive mechanisms to specifying the context information that can be consumed by services are thus needed. In this paper, and in order to bridge the gap of the context information perception by the developers and by the users, we propose to represent the available context information in each domain as a feature model. In this way, the developers are able to configure this feature model to get the context information they need, the users can easily specify the context features they are willing to share. The result is a domain-oriented user-centric privacy protection scheme.