Applying Chaos Theory for Hardware Trojan Detection
Hardware Trojans (HTs) pose a serious threat to the security of Integrated Circuits (ICs). Detecting HTs in an IC is an important but difficult problem due to the wide spectrum of HTs and their stealthy nature. A runtime Trojan detection system could monitor an IC during its operational life time and provide a last-line of defense. In this talk, we discuss a hardware-based runtime detection approach. It applies chaos theory to characterize dynamic power consumption data in a reconstructed phase space. The proposed runtime approach does not make any assumption on the statistical distribution of power consumption, thus making it applicable for runtime use. Hardware overhead, which is the main challenge for runtime approaches, is reduced by taking advantage of available thermal sensors present in most modern ICs. For real world implementation, thermal sensor noise is also discussed. The simulation results for detecting Trojans on publicly available Trojan benchmarks demonstrate that the proposed approach outperforms the current runtime Trojan detection approaches in terms of detection rate, computational complexity, and implementation feasibility.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
- Fairleigh Dickinson University
- Teaneck, New Jersey
- United States 07666
- Building: Auditorium M105, Muscarelle Center
- Click here for Map
- Contact Event Host
-
Alfredo Tan, tan@fdu.edu, Howard Leach h.leach@ieee.org
- Co-sponsored by SP01 and School of Computer Sciences and Engineering, FDU, AP/MTT
- Starts 01 April 2020 02:00 PM UTC
- Ends 22 April 2020 06:00 PM UTC
- No Admission Charge
Speakers
Dr. Hong Zhao of Fairleigh Dickinson University
Applying Chaos Theory for Hardware Trojan Detection
Biography:
Dr. Hong Zhao received Ph. D from New Jersey Institute of Technology in Electrical and Computer Engineering. She is a Professor of Electrical and Computer Engineering at Fairleigh Dickinson University, New Jersey, US. Her research focuses on various aspects of broadband communications and computer security including Network Traffic/Performance/Security Analysis and Modeling, and Hardware Trojan detection. She serves as Associate Editor of the Journal on Multidimensional Systems and Signal Processing, and Editor of the Journal of Computing and Information Technology. Professor Zhao also serves as a Vice Chair of IEEE North Jersey Section. She has been a TPC chair/member, symposium chair, and technical paper reviewer for IEEE conferences, journal magazines. Dr. Zhao served as a board Chair of the Wireless and Optical Communication Conference (WOCC) (2017-2018). She received AFRL VFRP award in 2014-2016, AFOSR SFFP award in 2017-2019, Visiting Professor Award from Ministry of Science and Technology Taiwan in 2015, and 2015 IEEE Region 1 award for Outstanding Support for the Mission of the IEEE, MGA, REGION 1 and Section.
Agenda
Hardware Trojans (HTs) pose a serious threat to the security of Integrated Circuits (ICs). Detecting HTs in an IC is an important but difficult problem due to the wide spectrum of HTs and their stealthy nature. A runtime Trojan detection system could monitor an IC during its operational life time and provide a last-line of defense. In this talk, we discuss a hardware-based runtime detection approach. It applies chaos theory to characterize dynamic power consumption data in a reconstructed phase space. The proposed runtime approach does not make any assumption on the statistical distribution of power consumption, thus making it applicable for runtime use. Hardware overhead, which is the main challenge for runtime approaches, is reduced by taking advantage of available thermal sensors present in most modern ICs. For real world implementation, thermal sensor noise is also discussed. The simulation results for detecting Trojans on publicly available Trojan benchmarks demonstrate that the proposed approach outperforms the current runtime Trojan detection approaches in terms of detection rate, computational complexity, and implementation feasibility.