Skip to content
Kean CS Lab

Program 03 / 06

Security, Privacy & Trust Architectures

Security is not a perimeter. It is a set of assumptions you can state — and a set you can test.

Faculty work ranges from cybersecurity and software assurance to privacy-preserving sensing and zero-trust designs for smart environments. The Ph.D. program names cybersecurity as a priority domain. This laboratory treats that as a research area: threat models, measurements, and systems you can evaluate in a setting you control — not a certification checklist.

Questions the work is for

  1. 01

    What does zero trust look like when the ‘enterprise’ is a home full of sensors?

  2. 02

    How do we learn from mobile and IoT traces without reconstituting the person?

  3. 03

    Where does explainability help a defender — and where does it help an adversary?

Methods, data, benchmarks

method

Zero-trust & explainable security

Public work with Kwak on privacy-preserving smart-home security using zero-trust architecture and explainability — a template for systems papers that name both the control and the human-readable rationale.

dataset

Mobile sensing traces

On-device and participatory sensing (Hu) with explicit consent, minimization, and attack-surface notes. We do not treat a smartphone as a free sensor.

method

Privacy threat modeling

STRIDE-style and LINDDUN-style analysis before a line of training code. If you cannot name the adversary, you do not yet have a security thesis.

tooling

Embedded ML security

Model extraction, membership inference, and side channels on constrained devices (Li) — security properties of the artifact, not only of the network.

Taxonomy on this page: method, dataset, benchmark, tooling. Field-standard methods are named as such; they are not automatic claims of a Kean dataset release.

Doctoral tracks

Smart environments

Kwak’s IoT and CPS work, plus the smart-home security paper, give a concrete setting: homes, campuses, and civic sensors that cannot assume a trusted LAN.

Privacy in learning

Federated and privacy-aware orchestration appears in Kwak’s 6G/consumer-service papers. Doctoral projects can take the measurement seriously: what leakage remains after the protocol?

Cybersecurity education & practice

Li, Steiner, and Morreale (WiCyS) connect security research to how the next cohort is trained. A methods contribution can be a lab, a dataset, or a curriculum instrument — if it is evaluated.

Who this is for

Come with systems or security coursework and a habit of writing threat models. We are not looking for a survey of tools; we are looking for a claim you can falsify.

Ph.D. Program →

Faculty in this program

Selected papers

← All programs