Mobile systems, such as smartphones, wearables (e.g., smart watches, AR/VR headsets), and IoT devices, have evolved significantly from being just a method of communication to sophisticated sensing devices that monitor and control several aspects of our lives. While these devices' advanced computing and sensing capabilities enable several useful applications, they also make them attractive targets for attackers, leading to several security threats and loss of privacy.
In this talk, I explore how we can leverage multimodal sensing to design secure and usable user and device authentication systems for mobile and IoT devices through a combination of computer vision, machine learning, and cryptographic methods. In particular, I will present two systems for mobile security and privacy: a liveness detection system for protecting face authentication systems on mobile devices, and a group pairing system for IoT devices with different sensing modalities. Through this research, we develop tools and algorithms that allow developers to implement effective and user-friendly systems for user and device authentication.
Habiba Farrukh is a PhD candidate in the Computer Science department at Purdue University, where she is advised by Professor Z. Berkay Celik. Habiba has conducted research on a variety of topics, including mobile and IoT security and privacy and human-centered computing. Her dissertation focuses on leveraging multimodal sensing on mobile and IoT devices to provide rigorous security and privacy guarantees for these systems. She received the Bilsland Dissertation Fellowship for her dissertation in 2021. She has led a team of 5 students to improve the usability and conformity of Android authorization APIs and WearOS permission model for the Google ASPIRE (Android Security and PrIvacy REsearch) projects in 2021 and 2022. She is expected to earn her PhD in the Spring of 2023. Habiba also interned with the Machine Learning Science team at Amazon Robotics.