With the proliferation of computing systems in our world, from servers to internet-of-things devices, side-channel signals have become significantly more available and accessible to measure and leverage. This availability provides both opportunities and security threats. On one hand, these side-channels may “leak” sensitive information about the system, and if exploited by an adversary, it would pose security threats to the system. On the other hand, however, these signals can be leveraged as extra sources of information which can be used for benign and useful purposes such as debugging/profiling, malware/intrusion detection, and security monitoring. Given these opportunities and threats, understanding how these side-channels are created and developing frameworks to leverage them is an important topic.

To address these challenges, in this talk, I will describe our new methods and frameworks to identify, quantify, and leverage side-channel signals particularly analog-domain electromagnetic (EM) emanations, and then describe how these analog-domain side-channels can be leveraged for intrusion detection on a variety of embedded and cyber-physical systems. Moreover, I will present our recent work on how side-channels can be leveraged to create a trusted execution environment (TEE) on a resource-constrained device (e.g., an Internet-of-Things device).