Key research themes
1. How can timing analysis of inter-signal arrival times detect regular-frequency signal injection attacks on Controller Area Network (CAN) buses in vehicles?
This research area focuses on leveraging the regularity and predictability of CAN bus signal frequencies to detect injection attacks. Since injected commands must maintain regular, redundant timing to affect vehicle behavior, anomalies in the timing patterns—specifically inter-signal wait times—can reveal intrusions. Detection mechanisms that model these timings provide a scalable defensive approach that does not depend on reverse engineering specific vehicle signal mappings.
2. What are the advances in signal injection coding to achieve spectral nulls for improved communication system spectral shaping?
This theme encompasses the design of injection and permutation coding schemes coupled with pulse amplitude modulation (PAM) to shape the spectral content of transmitted signals, particularly to place spectral nulls at designated rational sub-multiples of the symbol frequency. Achieving spectral nulls optimizes bandwidth utilization and reduces interference in digital communication and storage systems, such as optical fibers and LANs.
3. How can fault injection leveraging FPGA partial reconfiguration accelerate fault tolerance validation for integrated circuits?
This research investigates emulation-based fault injection strategies on FPGAs using dynamic partial reconfiguration (DPR) to efficiently inject faults without intrusive hardware modifications. The objective is to accelerate large-scale fault campaigns, improve fault coverage, and enable earlier validation in IC design by emulating radiation-induced transient errors like single event upsets.
4. What insights have recent studies provided into control logic injection attacks targeting programmable logic controllers (PLCs), and what are their primary attack modalities?
This theme covers comprehensive analyses of control logic injection attacks on PLCs that threaten critical infrastructure by maliciously altering control program behavior. Researchers categorize these attacks into code modification, input manipulation, and memory corruption scenarios, each with distinct attack vectors and consequences. Understanding these categories informs the design of defensive and detection mechanisms.
5. How can signal injection be harnessed for enhanced state observation and control in electromechanical systems through virtual output generation?
This research investigates high-frequency signal injection into nonlinear electromechanical systems to create virtual outputs—high-frequency components in the measured output—that enable estimation of otherwise unobservable states. Development and application of innovative filtering techniques improve the extraction of these virtual outputs, facilitating observer design and sensorless control.
6. What advancements exist in injectable wireless microdevice technology, and what challenges affect their delivery and interfacing in biomedical applications?
Focusing on the miniaturization of fully injectable wireless implantable medical devices, this theme explores state-of-the-art micro/nanofabrication, powering, and communication techniques. Emphasis is placed on biomedical applications requiring minimally invasive implantation, chronic stability, reduced immune response, and distributed physiological interfacing, alongside the challenges in device encapsulation, injection methods, and multi-device networking.
7. How can timing- and frequency-based signal injection methods enhance rotor position self-sensing in electric drives?
This theme covers signal injection techniques in permanent magnet synchronous machines to estimate rotor position without mechanical sensors. By injecting high-frequency signals and analyzing anisotropic machine responses, researchers improve low-speed and standstill position estimation, addressing challenges from electromagnetic interference and anisotropic parameter variations.
8. What fault injection methods are effective for analyzing bidirectional digital signals, and what challenges do such signals pose for fault injection testing?
Research here addresses how to inject faults in digital circuits with bidirectional I/O ports, where traditional fault injection must consider not only value corruption but also direction-related faults. It explores injection system design minimizing intrusion and preserving original netlist structures to enable reliable testing of microprocessor and peripheral designs in fault tolerance evaluation.
9. How does electromagnetic (EM) pulse injection create faults in integrated circuits, and what fault model best explains their occurrence?
This area focuses on understanding the fault mechanisms induced by electromagnetic pulse injection in ICs. Recent studies challenge the traditional view of EM faults as timing faults and instead present a model describing them as sampling faults in flip-flops due to disrupted sampling processes. The spatial locality and nature of these faults inform developing more effective fault injection attacks and countermeasures.
10. How can coding of sensor outputs enhance detection of stealthy false data injection attacks in cyber-physical systems?
This research focuses on applying coding matrices to sensor measurements to increase estimation residues and thus trigger anomaly detectors even under sophisticated attacks that are otherwise stealthy to conventional monitors. By mathematically designing these coding schemes, the system security against intelligent attackers who know the system model but not the coding matrix can be improved with relatively low computational overhead compared to full encryption.
