Date of Graduation

5-2022

Document Type

Thesis

Degree Name

Master of Science in Computer Science (MS)

Degree Level

Graduate

Department

Computer Science & Computer Engineering

Advisor/Mentor

Panda, Brajendra N.

Committee Member

Nakarmi, Ukash

Second Committee Member

Zhang, Lu

Keywords

cyber attacks; infrastructure; simulation

Abstract

Critical infrastructure systems have recently become more vulnerable to attacks on their data systems through internet connectivity. If an attacker is successful in breaching a system’s defenses, it is imperative that operations are restored to the system as quickly as possible. This thesis focuses on damage assessment and recovery following an attack. A literature review is first conducted on work done in both database protection and critical infrastructure protection, then the thesis defines how damage affects the relationships between data and software. Then, the thesis proposes a model using a graph construction to show the cascading affects within a system after an attack. This thesis also presents an algorithm that uses the graph to compute an optimal recovery plan that prioritizes the most important damaged components first so that the vital modules of the system become functional as soon as possible. This allows for the most critical operations of a system to resume while recovery for less important components is still being performed. The thesis shows results from simulations using the recovery algorithm on data graphs with various parameters. After that, a second model is proposed that accounts for the time elapsed after an attack to perform a more precise damage assessment. By doing this, it can be determined how far damage can spread, then unaffected parts of the system can be released for possible use. Simulations are also done on this model to show the changes in damage assessment when different parameters are altered.

Download

Share

COinS