DongInn Kim

About Me

Hi! My name is DongInn Kim. I am a PhD candidate in Computer Science at Indiana University. My current research focuses on the reverse engineering for the security of IoT devices and building an intelligent dynamic analysis tool on the top of existing static/dynamic analysis security tools. My PhD thesis under the supervision of Prof. Jean Camp is based on developing an intrusion detection system by fingerprinting the accessing edge services and IoT devices with machine learning processes of the collected features. I work for the IU CSI CTF (Capture The Flag) team and I am also working on the Open Source Cluster Application Resources (OSCAR) as a core developer. My research interests include software engineering, clustering computing, and systems security.

IU CSI CTF: Capture The Flag team of Indiana University

Research Interest

My current research focuses on the reverse engineering for the security of IoT devices and building an intelligent dynamic analysis tool on the top of existing static/dynamic analysis security tools.

Current Work

work on the SCGSR Awards in PNNL for the Summer intern: the proposed project is to develop a system to detect the BGP hijacking in the IoT environment.
mentor an REU student, Eric Swiler for the project of the SSL certification database

Past Work

detect malicious domains from the CH registry database: planning to apply machine learning
binary analysis to investigate any possible penetrations in Intel SGX: working with SGX reading group
practice the CTF exercises / problems with some various reverse engineering tools (e.g. radare2, unicorn, capstone, keystone, …)
renovating the security toolbar, which is an extension of Firefox to make web browsing more secure.
work on the “CUTS CertProbe and CertWarehouse” for collecting and compiling the Public key certificates. <= [Previous work: PDF]

Publications

Vafa Andalibi, Jayati Dev, DongInn Kim, Eliot Lear, and Jean Camp. Making Access Control Easy in IoT. In IFIP International Symposium on Human Aspects of Information Security & Assurance, June 2021.
Vafa Andalibi, Eliot Lear, DongInn Kim, and Jean Camp. On the Analysis of MUD-Files’ Interactions, Conflicts, and Configuration Requirements Before Deployment. In 5th EAI International Conference on Safety and Security in Internet of Things, SaSeIoT, May 2021. Springer.
Shakthidhar Gopavaram, Jayati Dev, Marthie Grobler, DongInn Kim, Sanchari Das, and L Jean Camp. Cross-National Study on Phishing Resilience. In Proceedings of the Workshop on Usable Security and Privacy (USEC), May 2021. => PDF
DongInn Kim, Vafa Andalibi, and L Jean Camp. Protecting IoT Devices through Localized Detection of BGP Hijacks for Individual Things. In SafeThings 2021, Oakland, May 2021. IEEE Workshop on the Internet of Safe Things. => PDF
DongInn Kim, Vafa Andalibi, and L Jean Camp. Fingerprinting Edge and Cloud Services in IoT. In Systematic Approaches to Digital Forensic Engineering, City University of New York (CUNY), New York City, May 2020. IEEE Computer Society. => PDF
Vafa Andalibi, DongInn Kim, and L. Jean Camp. Throwing MUD into the FOG: Defending IoT and Fog by expanding MUD to Fog network. In 2nd USENIX Workshop on Hot Topics in Edge Computing (HotEdge 19), Renton, WA, July 2019. USENIX Association. => PDF
DongInn Kim, Jeffrey M. Squyres, and Andrew Lumsdaine: The Introduction of the OSCAR Database API (ODA) => PDF
DongInn Kim, Jeffrey M. Squyres, and Andrew Lumsdaine: Revamping the OSCAR database: A Flexible Approach to Cluster Configuration Data Management. => PDF

Technical Skills

Machine learning (ML) models that I have implmented with R:

Name	What does it do
K-Means	Unsupervised, non-parametric ML to classify data with the given K value
E-M	Unsupervised, parametric ML to find maximum likelihood or maximum a posteriori
Linear Regression	Supervised, parametric ML to find the linear model between given data sets (X,…, Y)
Logistic Regression	Supervised, parametric ML to find the categorical model between dependent variables
KNN	Unsupervised, non-parametric ML for classification and regression
Naive Bayes	Semi-supervised, parametric ML for the probabilistic classifier with the assumption of IID

Here is the list of the tools that I am currently learning:

Name	What does it do
Frida	Dynamic instrumentation toolkit for the binary analysis
Scapy	Python version of tshark and more
Radare2	Reverse engineering framework
Unicorn	CPU emulator (U)
Capstone	Disassembly framework
Keystone	Assembler framework
Pwndbg	GDB plug-in to make debugging with GDB much easier

Research Colleagues

L Jean Camp: Professor, School of Informatics and Computing, Indiana University
Andrew Lumsdaine: UW-PNNL Distinguished Faculty Fellow and Affiliate Professor of Computer Science, University of Washington