AI and agentic systems for cyber-physical control

When an AI system can move something physical, what keeps it inside the envelope?

Machine learning is now embedded in decisions that open breakers, route traffic, and provision networks. Accuracy on a benchmark says little about behavior during an attack, a sensor fault, or a distribution shift. The center works on models that respect the physics of the system they control and on enforcement that sits between a model's output and the actuator.

What the group builds

Learning-based intrusion and anomaly detection

Scalable, real-time detection of attacks on grid control traffic, including adaptive transfer learning so a detector trained on one network still works on the next.

Federated learning across operators

Utilities cannot share raw operational data. Federated training lets detection models improve across sites while the measurements stay home, and the group measures what that costs in accuracy.

Physics-grounded models

A model that ignores power flow or optical impairment will confidently propose something impossible. The group grounds learned models in the physical constraints of the system.

Safety enforcement for agentic systems

As AI agents take actions rather than make predictions, the question becomes what the agent is permitted to do. The group works on enforcement layers that check actions against safety properties before they reach infrastructure.

Recent papers

The newest work from the faculty on this thrust. 72 papers since 2019 carry one of their names.

A. Rezaee, F. Arpanaei, R. McCann, H. Rabbani, J. A. Hernández, M. Brandt-Pearce, V. M. Vokkarane
QoT-Aware Dynamic Resource Allocation and Grooming in Multi-Band Space-Division Multiplexing Networks
IEEE/Optica Journal of Optical Communications and Networking, vol. 18, no. 10, Oct. 2026IF 4.0 (2023)
C. Pozzi, C. Ng, S. Lyon, Y. Luo, C. Niezrecki, M. Inalpolat
A Novel Distributed Sensor Network to Assess Wind Turbine Blade Acoustics for Structural Health Monitoring
Wind Energy, vol. 29, no. 10, Sept. 2026IF 4.1 (2023)
A. Rezaee, R. McCann, V. M. Vokkarane
FUSION: A Unified Benchmarking Framework for Reproducible Optical Network Research
IEEE/Optica Journal of Optical Communications and Networking, vol. 18, no. 9, pp. D90-D105, Sept. 2026 (Special Issue on Benchmarking in Optical Networks)IF 4.0 (2023)
A. Rezaee, F. Arpanaei, R. McCann, L. Nadal, J. A. Hernández, V. M. Vokkarane
QoT-Aware Spectral and Spatial Scaling Trade-offs in Multi-Band Space Division Multiplexing over EONs [Invited]
IEEE/Optica Journal of Optical Communications and Networking, vol. 18, no. 8, pp. C160-C172, Aug. 2026IF 4.0 (2023)
T. Miskell, Y. Luo, P. Li, S. W. Lim
LLM Based Static Malware Analysis for Windows Portable Executables
35th International Conference on Computer Communications and Networks (ICCCN), pp. 1-6, July 2026
H. Rabbani, A. Rezaee, H. Rabbani, V. M. Vokkarane, M. Brandt-Pearce
Experimental Determination of Filter Bandwidth Requirements for Coherent Pluggable Transceivers in Optical Data Center Networks
IEEE International Conference on High Performance Switching and Routing (HPSR), pp. 1-5, June 2026
M. Z. Islam, Y. Lin, V. M. Vokkarane
Disaster-Resilient Cyber-Physical Distribution System Reconfiguration and Dynamic Networked Microgrid Formation Under Intermittent Generation
IEEE Transactions on Industry Applications, vol. 62, no. 2, pp. 3459-3471, Mar. 2026IF 4.2 (2023)
S. Lyon, C. A. Ng, C. Pozzi, M. Inalpolat, C. Niezrecki, Y. Luo
Signal Strength and Network Performance Optimization of a Wireless Acoustic Sensor for Wind Turbine Blade Health Monitoring
IEEE Sensors Journal, vol. 26, no. 3, pp. 5195-5203, Feb. 2026IF 4.3 (2023)

All publications

Other thrusts

Most projects cut across two or three of them.

Smart grid cybersecurity and resilience Next-generation optical and 6G transport Fault-tolerant distributed and edge computing Hardware security and high performance computing Connected transportation, health, and infrastructure Nuclear energy and security