Research

My dissertation ties together four studies that look unrelated but answer the same question: IoT defenses today are more complex than they need to be, built in isolation from one another, and able to flag an attack without explaining it.

Interests

IoT and embedded securityMachine learning for securityInterpretability and visual analyticsData provenance and integrityHigh-performance computing and benchmarkingTooling for data-intensive research

How it fits together

Three problems

Complexity by defaultProposed defenses routinely exceed what real IoT hardware can carry
Defense stages in isolationAuthentication, detection, and response are designed separately and never share what they learn
Detection without interpretationA model flags an anomaly but offers nothing an analyst can act on

Four studies

Provenance authenticationVerifying where IoT data came from, at a cost small devices can bear
IoT device identificationIdentifying device type and identity from communication behavior
Crypto ransomware detectionDetecting file-encrypting ransomware from system behavior traces
IoT attack behavior visualizationMaking attack behavior visible so an analyst can reason about it

One dissertation

Data-Driven Cybersecurity for Connected Devices: Traffic, Provenance, and Behavioral Threat Detection
The four studies answer all three problems together — the diagram does not claim one study per problem.

Publications

Eight peer-reviewed papers, first author on every one: six full papers and two short papers.

2026
  1. A Web-Based Interactive Visualization Tool for Astronomical Time Series
    Roemsri, P., McCarver, A., Maccarone, T., Shaw, A. W., Gandhi, P., Kaplan, D., & Dang, T. · AIP Conference Proceedings, 3378(1), 040001
2025
  1. PowerMap and Time-Indexed PCA Trajectories: Interactive Visual Forensics for HPC Incidents
    Roemsri, P., Ribeiro, A. H., & Dang, T. · CIDDF 2025 — Advances in Computer Science Applications and Research, Vol. 5, SpringerDOI
2023
  1. Visualization of Attack Behavior Data in IoT Network Traffic
    Roemsri, P., & Dang, T. · 2023 5th International Conference on Big-Data Service and Intelligent Computation (BDSIC), pp. 1–8
  2. On Detecting Crypto Ransomware Attacks: Can Simple Strategies be Effective?
    Roemsri, P., Puangpontip, S., & Hewett, R. · 6th International Conference on Information and Computer Technologies (ICICT), pp. 138–143DOI
  3. An Interactive Multi-Degree Analysis Approach Short paper
    Roemsri, P., & Dang, T. · IEEE VAST Challenge, pp. 9–10
  4. Effective Dynamic Network Layout for Transaction Analysis Short paper
    Roemsri, P., & Dang, T. · IEEE VAST Challenge, pp. 7–8
2021
  1. Device Identification for IoT Security
    Roemsri, P., & Hewett, R. · IEEE 6th International Conference on Signal and Image Processing (ICSIP), pp. 866–870
2020
  1. Provenance Location-Based Authentication in Cloud Computing
    Roemsri, P., & Hewett, R. · 11th International Conference on Advances in Information Technology (IAIT), Article 24, pp. 1–8DOI

Awards

  • Best Presentation, Session 3 — APCT 2025 — 2025 4th Asia-Pacific Computer Technologies Conference, Bangkok, January 2025. For A Web-based Interactive Visualization Tool for Astronomical Time Series.
  • The Engineering Student Talent Award, first runner-up — June 2017. Selected to one of the teams representing Thailand at the ASEAN programme in India.