Search for: All records

A smart home involves a variety of entities, such as IoT devices, automation applications, humans, voice assistants, and companion apps. These entities interact in the same physical environment, which can yield undesirable and even hazardous results, called IoT interaction threats. Existing work on interaction threats is limited to considering automation apps, ignoring other IoT control channels, such as voice commands, companion apps, and physical operations. Second, it becomes increasingly common that a smart home utilizes multiple IoT platforms, each of which has a partial view of device states and may issue conflicting commands. Third, compared to detecting interaction threats, their handling is much less studied. Prior work uses generic handling policies, which are unlikely to fit all homes. We present IoTMediator, which provides accurate threat detection and threat-tailored handling in multi-platform multi-control-channel homes. Our evaluation in two real-world homes demonstrates that IoTMediator significantly outperforms prior state-of-the-art work. 

Implicit authentication for traditional objects, such as doors and dumbbells, has rich applications but is rarely studied. An ongoing trend is that traditional objects are retrofitted to smart environments; for instance, a contact sensor is attached to a door to detect door opening (but cannot tell "who is opening the door"). We present the first accurate implicit-authentication system for retrofitted everyday objects, named MoMatch. It makes an authentication decision based on a single natural object use, unlike prior work that requires shaking objects. MoMatch is built on the observation that an object has a motion typically because a human hand moves it; thus, the object's motion and the legitimate user's hand movement should correlate. The main challenge is, given the small amount of data collected during one object use, how to measure the correlation accurately. We convert the correlation measurement problem into an image comparison problem and resolve it using neural networks successfully. MoMatch does not need to profile the user's biometric information and is resilient to mimicry attacks. 

As Internet of Things (IoT) technology becomes more widespread and commonplace in homes, the efficiency of these devices using available bandwidth is becoming more of a concern, as the number of connected devices in a home increase drastically. If each device is controlled using a separate Application Programming Interface (API), the strain on a network will be much worse than it would if all these devices are controlled from a single point. This single point could handle all commands to and from the devices, thereby decreasing the network load. The framework of a testbed presented in this paper will allow developers to build an API around the devices included in the testbed. Then test their algorithms and other research methods from a remote location.