Abstract

As the world grows more and more reliant on microcomputing systems, security becomes a greater and greater concern. Malicious code is an ever-present issue in the field of computing and hardware vulnerabilities pose a rising threat to an increasingly connected landscape. This variety in threats can be mitigated through invariant rules for device operation, which form a policy for acceptable device behavior such that policy violations can be interpreted as either erroneous or malicious. This defense architecture disregards threat source and identifies issues based on their symptoms. Such systems are usually enforced through hardware implementations, which can become expensive in time and resources to include and even more so to adapt or modify. To address this, a software-based anomaly detector system was designed and implemented on an STM32-l476rg microcontroller. This system enforces a set of custom security invariants through software and included device peripherals, allowing anomaly detection without modifying the device. These invariants were developed over the course of the work based on listings from the Common Weakness Enumeration (CWE) from the MITRE corporation. The detector consists of a secure bootloader to ensure code integrity, a detector system making use of general-purpose and watchdog timers, and a FreeRTOS instance to run the overlying user code. This system was tested against simulated hardware anomalies through the debug port and measured for performance. The detector was able to successfully detect a variety of threats, including out-of-bounds execution, locked memory modification, stack overflow, improper state transition, hardware reconfiguration, debug mode use, secure boot bypass, and AES round skips. The detector code was found to have a processor time overhead of 0.0842 percent, and the memory usage reached about four percent utilization for the target processor.

Library of Congress Subject Headings

Computer security--Software; Computers--Defects; Debugging in computer science

Publication Date

12-2023

Document Type

Thesis

Student Type

Graduate

Degree Name

Computer Engineering (MS)

Department, Program, or Center

Computer Engineering

College

Kate Gleason College of Engineering

Advisor

Michael Zuzak

Advisor/Committee Member

Roy Melton

Advisor/Committee Member

Marcin Lukowiak

Comments

This thesis has been embargoed. The full-text will be available on or around 12/22/2024.

Campus

RIT – Main Campus

Plan Codes

CMPE-MS

Download

Available for download on Saturday, December 21, 2024

Share

COinS