Date of Award

3-14-2014

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Operational Sciences

First Advisor

Jennifer L. Geffre, PhD.

Abstract

SCADA (supervisory control and data acquisition) systems monitor and control many different types of critical infrastructure such as power, water, transportation, and pipelines. These once isolated systems are increasingly being connected to the internet to improve operations, which creates vulnerabilities to attacks. A SCADA operator receives automated alarms concerning system components operating out of normal thresholds. These alarms are susceptible to manipulation by an attacker. This research uses information theory to build an anomaly detection model that quantifies the uncertainty of the system based on alarm message frequency. Several attack scenarios are statistically analyzed for their significance including someone injecting false alarms or hiding alarms. This research evaluates the use of information theory for anomaly detection and the impact of different attack scenarios.

AFIT Designator

AFIT-ENS-14-M-32

DTIC Accession Number

ADA610092

Share

COinS