Purdue University Graduate School
Browse
Gihan_Mendis_Dissertation.pdf (5.32 MB)

DEEP LEARNING FOR SECURING CRITICAL INFRASTRUCTURE WITH THE EMPHASIS ON POWER SYSTEMS AND WIRELESS COMMUNICATION

Download (5.32 MB)

Imbulgoda Liyangahawatte, Gihan Janith Mendis Ph.D., Purdue University, May

2023. Deep learning for securing critical infrastructure with the emphasis on power

systems and wireless communication. Major Professor: Dr. Jin Kocsis.


Critical infrastructures, such as power systems and communication

infrastructures, are of paramount importance to the welfare and prosperity of

modern societies. Therefore, critical infrastructures have a high vulnerability to

attacks from adverse parties. Subsequent to the advancement of cyber technologies,

such as information technology, embedded systems, high-speed connectivity, and

real-time data processing, the physical processes of critical infrastructures are often

monitored and controlled through cyber systems. Therefore, modern critical

infrastructures are often viewed as cyber-physical systems (CPSs). Incorporating

cyber elements into physical processes increases efficiency and control. However, it

also increases the vulnerability of the systems to potential cybersecurity threats. In

addition to cyber-level attacks, attacks on the cyber-physical interface, such as the

corruption of sensing data to manipulate physical operations, can exploit

vulnerabilities in CPSs. Research on data-driven security methods for such attacks,

focusing on applications related to electrical power and wireless communication

critical infrastructure CPSs, are presented in this dissertation. As security methods

for electrical power systems, deep learning approaches were proposed to detect

adversarial sensor signals targeting smart grids and more electric aircraft.

Considering the security of wireless communication systems, deep learning solutions

were proposed as an intelligent spectrum sensing approach and as a primary user

emulation (PUE) attacks detection method on the wideband spectrum. The recent

abundance of micro-UASs can enable the use of weaponized micro-UASs to conduct

physical attacks on critical infrastructures. As a solution for this, the radio

frequency (RF) signal-analyzing deep learning method developed for spectrum

sensing was adopted to realize an intelligent radar system for micro-UAS detection.

This intelligent radar can be used to provide protection against micro-UAS-based

physical attacks on critical infrastructures.

History

Degree Type

  • Doctor of Philosophy

Department

  • Technology

Campus location

  • West Lafayette

Advisor/Supervisor/Committee Chair

Dr. Jin Kocsis

Additional Committee Member 2

Dr. Arjuna Madanayake

Additional Committee Member 3

Dr. Baijian Yang

Additional Committee Member 4

Dr. Julia M. Rayz