Files in this item

FilesDescriptionFormat

application/pdf

application/pdfHIRZALLAH-DISSERTATION-2018.pdf (1MB)
(no description provided)PDF

Description

Title:Security of cyber-physical systems: A control-theoretic perspective
Author(s):Hirzallah, Nabil H. K.
Director of Research:Voulgaris, Petros G.
Doctoral Committee Chair(s):Voulgaris, Petros G.
Doctoral Committee Member(s):Hovakimyan, Naira; Sauer, Peter W.; Salapaka, Srinivasa M.; Belabbas, Mohamed A.
Department / Program:Electrical & Computer Eng
Discipline:Electrical & Computer Engr
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Security, Control Systems
Abstract:Motivated by the attacks on control systems through the cyber (digital) part, we study how signal attacks injected through actuators and/or sensors affect control system stability and performance. We ask the questions: What are the different types and scenarios of signal attacks? When are the attacks stealthy and unbounded? How to compute the worst stealthy bounded attacks? How to defend against such attacks through controller design? How to identify and estimate signal attacks before significant performance loss happens? We answer the above questions in this thesis using tools from control theory. We show that it is necessary to use a sampled-data framework to accurately assess the vulnerabilities of control systems. In addition, we show that the most lethal attacks are related to the structure of the system (location of zeros and poles, number of inputs and outputs). We show that dual rate control is a powerful tool to defend against these vulnerabilities, and we provide a related controller design. Furthermore, we show that the worst stealthy bounded attacks can be computed by an iterative linear program, and we show how to lessen their effects through iterative controller design. Finally, we study the trade-off between control and estimation of signal attacks and provide several controller designs utilizing the power of dual rate sampling.
Issue Date:2018-11-15
Type:Thesis
URI:http://hdl.handle.net/2142/102434
Rights Information:Copyright 2018 Nabil Hirzallah
Date Available in IDEALS:2019-02-06
Date Deposited:2018-12


This item appears in the following Collection(s)

Item Statistics