Project Overview
Survival of the fittest and natural selection are concepts that can describe how species have developed better defenses (among other things) over time... so why not use these ideas for computer security?

This project aims to develop a new Moving Target (MT) defense strategy using a unique application of Genetic Algorithms (GAs) to manage computer configurations. The objective is to create an evolutionary-inspired system that proactively finds secure computer configuration postures while maintaining a desired level of diversity. The approach will increase the complexity and cost for attackers while reducing the exposure of vulnerabilities and increasing system resiliency.

What is a moving target defense and how can genetic algorithms be used?

Moving Target (MT) environments provide security through diversity by changing various system properties that are explicitly defined in the computer configuration. Temporal diversity can be achieved by making periodic configuration changes; however in an infrastructure of multiple similarly purposed computers diversity must also be spatial, ensuring multiple computers do not simultaneously share the same configuration and potential vulnerabilities. Given the number of possible changes and their potential interdependencies discovering computer configurations that are secure, functional, and diverse, is challenging.

A Genetic Algorithm (GA) can be employed to find temporally and spatially diverse secure computer configurations. In the proposed approach a computer configuration is modeled as a chromosome, where an individual configuration setting is a trait or allele. The GA operates by combining multiple chromosomes (configurations) which are tested for feasibility and ranked based on performance which will be measured as resistance to attack. Successive iterations of the GA yield configurations that are often more secure and diverse due to the crossover and mutation processes.

Project Meetings and Members
Meeting Time (Fall 2013): Wednesdays 12:00 - 1:00pm, Manchester Hall Room 17

Faculty: Errin Fulp, David John, William Turkett, Daniel Canas, and Don Gage
Students: Xin Zhou, Scott Seal, Matt McNiece, Kohler Kane, and John Passarelli
Graduated Students: Robert Smith and Brian Lucas

Interested? If you are interested in joining the project just email fulp @

Papers and Presentations
"Evolutionary Based Moving Target Cyber Defense." David J. John, Robert W. Smith, William H. Turkett, Daniel Canas, and Errin W. Fulp. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO) Workshop on Genetic and Evolutionary Computation in Defense, Security and risk management (SecDef), 2014.

"An Initial Framework for Evolving Computer Configurations as a Moving Target Defense." Brian Lucas, Errin W. Fulp, David J. John, and Daniel Canas. In Proceedings of the 9th Annual Cyber and Information Security Research Conference (CISRC) , 2014.

"An Automated System for Evolving Secure Systems." Brian F. Lucas, Masters Project, 2013.

"An Evolutionary-Inspired Approach for Moving Target Defenses." (poster) Errin W. Fulp, Daniel Canas, David J. John, and William H. Turkett. NSF SaTC Kickoff Meeting, 2012. [pdf]

"Improving the Diversity Defense of Genetic Algorithm-Based Moving Target Approaches." Michael B. Crouse, Errin W. Fulp, and Daniel Canas. In Proceedings of the National Symposium on Moving Target Research, 2012. [pdf]

"A Moving Target Environment for Computer Configurations Using Genetic Algorithms." Michael B. Crouse and Errin W. Fulp. In Proceedings of the 4th Symposium on Configuration Analytics and Automation (SafeConfig 2011), 2011. [pdf]