Date of Award

12-2019

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Engineering and Sciences

First Advisor

Carlos E. Otero

Second Advisor

Ersoy Subasi

Third Advisor

Samuel P. Kozaitis

Fourth Advisor

Philip Bernhard

Abstract

As the Internet of Things revolution continues to become more prevalent in humanity's daily routine, securing these devices is paramount. Society has seen a substantial increase in activity in the cyber-warfare battle space, resulting in an increasing amount of security breaches every year. The responsibility of securing our devices can no longer rely solely on cyber-security engineers keeping systems hardened through Security Technical Implementation Guides and vulnerability scans; it must shift towards the developer. Previous research has been done in this area of securing our devices. However, these solutions rely heavily on cloud computing resources to perform computationally expensive algorithms. While these solutions do work, devices without the support of an extensive cloud backbone lack adequate protection through these methods. The goal of this thesis is to provide a modern approach through the use of machine learning and its incorporation into the typical design pattern, State Based Machines. The coupling of these two concepts can allow for an increased security posture at a small cost of overall system performance.

Download

Share

COinS