Practical Integrity Validation in the Smart Home with HomeEndorser

Modern smart home platforms facilitate home automation using trigger-action routines. While providing flexibility, routines may also cause serious threats to system integrity: untrusted third-parties may use platform APIs to modify the abstract home objects (AHOs) that high-integrity devices (e.g., security camera) rely on (i.e., as triggers). As most AHO accesses are legitimate, applying naive information flow controls or removing permissions would not only fail to prevent these problems, but also break functionality. Therefore, this paper proposes the alternate approach of home abstraction endorsement, which endorses a proposed AHO change by correlating it with expected environmental changes. We present the HomeEndorser framework, which provides a policy model to express changes in device states as endorsement policy templates that are automatically instantiated in a given configuration (based on device availability/placement), and a platform-based reference monitor to mediate all API requests to change AHOs. We implement HomeEndorser as an enhancement to the HomeAssistant platform, and demonstrate less than 10% performance overhead and no false alarms under realistic usage, as well as derive policy templates for 6 key AHOs.