Doctoral Dissertations

Orcid ID

0000-0002-6909-1022

Date of Award

5-2021

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Computer Engineering

Major Professor

Scott I. Ruoti

Committee Members

Kent Seamons, Jinyuan Sun, Doowon Kim, Scott I. Ruoti

Abstract

Security experts recommend password managers to help users generate, store, and enter strong, unique passwords. Prior research confirms that managers do help users move towards these objectives, but it also identified usability and security issues that had the potential to leak user data or prevent users from making full use of their manager. In this dissertation, I set out to measure to what extent modern managers have addressed these security issues on both desktop and mobile environments. Additionally, I have interviewed individuals to understand their password management behavior.

I begin my analysis by conducting the first security evaluation of the full password manager lifecycle (generation, storage, and autofill) on desktop devices, including the creation and analysis of a corpus of 147 million generated passwords. My results show that a small percentage of generated passwords are weak against both online and offline attacks, and that attacks against autofill mechanisms are still possible in modern managers. Next, I present a comparative analysis of autofill frameworks on iOS and Android. I find that these frameworks fail to properly verify webpage security and identify a new class of phishing attacks enabled by incorrect handling of autofill within WebView controls hosted in apps. Finally, I interview users of third-party password managers to understand both how and why they use their managers as they do. I find evidence that many users leverage multiple password managers to address issues with existing managers, as well as provide explanations for why password reuse continues even in the presence of a password manager. Based on these results, I conclude with recommendations addressing the attacks and usability issues identified in this work.

Comments

Final version after defense with corrections

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS