Date of Award

6-18-2015

Document Type

Thesis

Degree Name

Master of Science in Cyber Operations

Department

Department of Electrical and Computer Engineering

First Advisor

Brett J. Borghetti, PhD.

Abstract

Methods of performing and defending against deceptive actions are a popular field of study in game theory; however, the focus is mostly on action deception in turn-based games. This work focuses on developing strategies for performing environmental deception in two-player, strategic-form games. Environmental deception is defined as deception where one player has the ability to change the other's perception of the state of the game through modification of their perception of the game's payoff matrix, similar to the use of camouflage. The main contributions of this research are an expansion of the definition of the stability of a Nash equilibrium to include cells outside the equilibrium, and the creation of four algorithms for developing strategies for environmental deception, including closed-form solutions for the creation of a 3x3 deceptive game with a 2x2 mixed-strategy Nash equilibrium (MSNE) that benefits the deceiver from a 3x3 game containing a 2x2 MSNE. It is found that the value gain produced by a deceptive algorithm is dependent upon the type of game to which it is applied and the maximum amount of allowable change to the payoff matrix emphasizing the importance of carefully selecting an algorithm to match the situation to which it is applied.

AFIT Designator

AFIT-ENG-MS-15-J-004

DTIC Accession Number

ADA623164

Download

Share

COinS