HomePad: A Privacy-Aware Smart Hub for Home Environments

The adoption of smart home devices is hindered today by the privacy concerns users have regarding their personal data. Since these devices depend on remote service providers, users remain oblivious about how and when their data is disclosed and processed. In this paper we present HomePad, a privacy-aware smart hub for home environments. Our system aims to empower users with the ability to determine how applications can access and process sensitive data collected by smart devices (e.g., web cams) and to prevent applications from executing unless they abide by the privacy restrictions specified by the users. To achieve this goal, HomePad applications are implemented as directed graphs of elements, which consist of instances of functions that process data in isolation. By modeling elements and the flow graph using Prolog rules, HomePad allows for automatic verification of the application's flow graph against user-defined privacy policies. Homepad incurs a negligible performance overhead, requires a modest programming effort, and provides flexible policy support to address the privacy concerns most commonly expressed by potential smart device consumers.