Research Aim
Design and evaluate an OT sensor attack and fault simulation platform that generates realistic industrial sensor data and applies machine learning to classify abnormal behavior.
Research Story
The idea behind CyberSense AI Guard
The project focuses on distinguishing between normal sensor behavior, faults, and cyber-attack-driven manipulation in industrial environments.
Why the approach matters
The project emphasizes offline-first operation, real-time awareness, and dynamic analysis without requiring a cloud-dependent setup for demonstrations.
Project Journey
From monitoring to intelligent warning
1. Sensor data collection
Use simulated or live-style data streams to represent industrial sensor behavior.
2. Feature extraction
Process windows of readings to capture changes, patterns, and instability.
3. Threat and anomaly interpretation
Map suspicious patterns to known OT/ICS-oriented attack categories and risk states.
4. Alerting and explanation
Convert technical analysis into dashboard warnings, messages, and action-oriented awareness.