Fingerprinting

We show that multiple voice assistant platforms can be fingerprinting equally effectively. We also show that the fingerprinting process can be performed remotely mixed with traffic from other devices. Adding additional features such as flow and burst based features can also increase fingerprinting performance

We show that not only is it possible to effectively fingerprint 188 IoT devices (with over 97% accuracy), but also to do so even with multiple instances of the same make-and-model device. We also analyze the extent to which temporal, spatial and data-collection- methodology differences impact fingerprinting accuracy. Our analysis sheds light on features that are more ro- bust against varying conditions

In this poster, we try to understand how effectively an attacker can fingerprint unseen targeted IoT devices when building a classifier using either devices manufactured by the same company or devices with similar functionality

Understanding and reducing information leakage from IoT devices through network based channels.