Research Summaries

Back A Graded Approach to Moving Target Defense leveraging Partially Observable Markov Decision Processes

Fiscal Year 2019
Division Graduate School of Engineering & Applied Science
Department Electrical & Computer Engineering
Investigator(s) McAbee, Ashley S.
Sponsor Naval Information Warfare Center, Pacific (Navy)
Summary Executive Summary of Proposed Work. Taking our inspiration from nature, we propose a computer network defense system that makes graded dynamic defensive changes in response to an assessment of current network security risk similar to the way a moth makes different defensive decisions depending on its assessment of the danger of nearby predatory bats. While Moving Target Defense (MTD) appears promising toward improving cyber defense, specific implementation best practices are not well understood, particularly for systems that employ multiple MTD techniques. Based on our preliminary analysis, we assess that various tools including the hidden Markov model and partially observable Markov decision process (POMDP) could be leveraged to facilitate MTD implementation in optimal response to an automated assessment of current risk state. Our research will con- duct a qualitative study of the proposed system leveraging real-world data to estimate the probabilities needed to understand system effectiveness and efficiency. We will also conduct a small scale validation of the theory-based performance estimates via a small scale stand-alone network of virtualized resources and within network security simulation software.
Keywords Cyber Defense Moving Target Defense NIWC Fellowship
Publications Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal
Data Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal