Michigan State University, USA
Title: Backdoor Attacks Against Speaker Verification Systems
Speaker Verification (SV) systems have become integral in mobile systems and smart vehicles for authenticating users by their voice traits. In this talk, I will introduce the backdoor attack strategy designed to compromise these SV models. I will show that well-designed backdoor attacks could target scenarios where attackers have zero knowledge about the intended victim. I will delve into the constraints of existing poisoning attacks against unforeseen targets, optimizing a universal backdoor adept at assaulting any given target. A key feature of the backdoor attack is its ability to discreetly integrate the speaker’s unique traits and semantic details, rendering it undetectable. I will conclude the talk by discussing potential countermeasures.
Dr. Qiben Yan is an Assistant Professor in Department of Computer Science and Engineering of Michigan State University. He received his Ph.D. in Computer Science department from Virginia Tech. He was a security researcher in a cybersecurity startup company Shape Security, where he participated in building the first “botwall”. Before joining MSU, he was an Assistant Professor in the University of Nebraska-Lincoln. He is a recipient of NSF CRII award in 2016. He has won IEEE SECON 2021 and ACM SenSys 2021 best paper awards. His current research interests include IoT systems, blockchain system security, mobile and IoT security, and wireless communication and security. His research has been reported in various high-impact media outlets, including the BBC Radio, Scientific American, Science Daily, Forbes, Popular Mechanics, Gizmodo, The Register, etc. His recent research has been supported by NSF under the SaTC, CPS，NeTS, and SpecEES programs, DoE, and FORD.
University of Notre Dame, USA
Title: Attack Recovery for Cyber-Physical Systems
Cyber-physical systems (CPSs) rely on computing components to control physical objects, and have been widely used in real-world life-critical applications. However, a CPS has security risks by nature due to the integration of many vulnerable subsystems, which adversaries exploit to inflict serious consequences. Among various attacks, sensor attacks pose a particularly significant threat, where an attacker maliciously modifies sensor measurements to drift system behavior. There is a lot of work in sensor attack prevention and detection. Nevertheless, an essential problem is overlooked: recovery–what to do after detecting a sensor attack, which needs to safely and timely bring a CPS back. We aim to highlight the need to investigate this problem, outline its four key challenges, and provide a brief overview of initial solutions in the field.
Dr. Fanxin Kong is a tenure-track assistant professor in the Department of Computer Science and Engineering at the University of Notre Dame. Before that, Dr. Kong worked as a tenure-track assistant professor at Syracuse University and as a postdoctoral researcher at the University of Pennsylvania. He completed his Ph.D. in Computer Science at McGill University. Dr. Kong’s research addresses security, safety, real-time, and resource efficiency for Cyber-Physical Systems through machine learning techniques, formal methods, control, optimization, and algorithm design, with applications to autonomous systems such as self-driving cars and drones. He has published over 60 research papers at highly reputable venues, including RTSS, RTAS, EMSOFT, ICCPS, IoTDI, DAC, etc., various IEEE/ACM transactions, and books/book chapters. His research is supported by NSF, AFRL, AFOSR, DARPA, and has received multiple awards. He has served in organizing committees and technical program committees for many conferences and workshops. He is currently serving as the Information Director of ACM SIGBED.
Michigan Technological University, USA
Title: A Hierarchical Microchain Network Architecture for Next Generation Smart Vehicular Network
Thanks to the rapid advancements in Artificial Intelligence (AI), big data, and vehicular technologies atop 5G communication, an emerging concept of the Internet of Vehicles (IoV) has become realistic by providing diverse applications to realize the vision of intelligent transportation systems (ITSs).By enabling a common information exchange platform between connected smart vehicles and heterogeneous vehicular service networks, ITS aims to create a convenient and safe transport environment for all road participants. Meanwhile, interconnected vehicles and data exchange and storage among multiple service providers have also brought new concerns regarding performance, interoperability, security, and privacy. Due to its key characteristics of decentralization, tractability, and immutability, Blockchain is recognized as a promising technology to enhance security and privacy for IoV networks.
In this talk, we present a hierarchical microchain network architecture, which integrates a Blockchain federation paradigm with IoV networks to improve scalability, interoperability, and security in multi-domain smart vehicular systems. By leveraging a hierarchical microchain federation, each small-scale permissioned IoV domain network adopts a lightweight Microchain to ensure domain-specific performance, security, and privacy. While a public inter-microchain Blockchain interconnects fragmented microchains to improve scalability and guarantee the security of cross-domain operations at the global level.
Dr. Ronghua Xu is an Assistant Professor of Applied Computing at Michigan Technological University. He earned a Ph.D. and an M.S. in Electrical and Computer Engineering at the Binghamton University – State University of New York (SUNY) in 2023 and 2018, respectively. He also received an M.S. in Mechanical and Electrical Engineering from Nanjing University of Aeronautics & Astronautics in 2010 and a B.S. in Mechanical Engineering from Nanjing University of Science & Technology, China in 2007. Before joining Binghamton University, he worked at Siemens on software development, system integration, and test automation from June 2010 to June 2016. His research interests are Blockchain and Distributed Ledger Technology, Internet-of-Things (IoT), Machine Learning (ML), Network Security, and Next Generation Networks (NGNs).
Florida State University, USA
Title: Signal Emulation Attack and Defense in Heterogeneous IoT System
In the evolving landscape of the Internet of Things (IoT), safeguarding wireless security has become extremely crucial for emerging applications, such as wearable computing in healthcare, intelligent controls of home automation, and real-time machinery health monitoring for industrial IoT. These applications depend on diverse types of IoT data to support accurate decision-making, for which each wireless protocol plays a critical role in data collection in terms of data rate, communication range, and power consumption level. Current mainstream IoT protocols, such as Wi-Fi, ZigBee, and Bluetooth/Bluetooth Low Energy (BLE), heavily overlap on the Industrial, Scientific, and Medical (ISM) 2.4GHz bands, creating a heterogeneous wireless environment with a massive number of IoT devices. The prevalence of wireless coexistence among heterogeneous IoT devices over 2.4GHz opens the door to new security threats. While the network security issues of each protocol have been thoroughly investigated in both academia and industry, the security challenges arising from protocol coexistence – a fertile ground for potential signal emulation attacks – have rarely been discussed. This talk aims to spotlight these emerging security challenges and propose effective countermeasures to ensure secure data transmission in heterogeneous IoT systems.
Dr. Xiaonan Zhang is an assistant professor in the Department of Computer Science at Florida State University. She received her M.S. degree from the Department of Electrical and Computer Engineering at Binghamton University in August 2017 and her Ph.D. degree from the Department of Electrical and Computer Engineering at Clemson University in August 2020. Her research interests span over general areas of wireless communication and network, the Internet of Things (IoT), and wireless security. Her current research focuses on intelligent next-generation (NextG) wireless systems, with an emphasis on AI-enabled IoT, AI for future edge computing and distributed intelligence, and advanced communication paradigms. Her publications have appeared in prestigious conferences and journals, including ACM MobiHoc, IEEE INFOCOM, IEEE ICDCS, ACM AsiaCCS, IEEE TON, IEEE TMC, and IEEE TWC.