top of page
Anomaly Detection in BACnet systems
The Project
Our goal is to implement THE-driven semi-supervised Machine Learning (ML) techniques to allow for automated anomaly detection in BACnet traffic and ensure that this procedure alerts the user of an imminent attack on the hardware system.
​
In this project, we propose to generate large amounts of data by emulating a real BAN on Raspberry Pi stacks, so as to develop an anomaly detection method using semi-supervised machine learning. The anomalies could be synthetic attacks or malfunctions in BAN services. The focus is on reducing labeling work and improving accuracy by merging supervised and unsupervised techniques to create a unique program.
The Team
Sofian Ghazali
Rahul Balamurugan
Muhammad Zahid Kamil
Dr. Hussein Alnuweiri
Primary Faculty Advisor
Mr. Salah Hessein
Secondary Faculty Advisor
bottom of page