Team Members
Summary
Protecting the next-gen communication networks from failure and cyber attacks are of utmost priority to the network service providers. In this regard, predicting network failure and network intrusion is vital towards developing an intelligent preventive mechanism against such network failure and attacks. This research project will develop algorithms that can predict network failures and attacks (in an end-to-end packet-optical network) using state-of-the-art machine learning techniques. Besides, our model will dynamically reserve network resources to protect the network traffic subject to such potential failure or attack. These proposed predictive and resource-friendly dynamic allocation schemes will help the carriers protect their network traffic from physical failure and cyber attacks.
Publications
-
Towards a Novel Hybrid Intrusion Detection Framework using Deep Transfer Learning
submitted to IEEE Transactions on Network and Service Management [Impact Factor 3.878]
-
IEEE Network Intrusion Detection and Comparative Analysis using Ensemble Machine Learning and Feature Selection
submitted to IEEE Transactions on Networks and Service Management (TNSM) [Impact Factor: 3.878]
-
Some studies on the Optimum Feature Set for DDoS Classification
submitted to Elsevier Computer Networks [Impact Factor: 4.20]
-
Extracting Optimal Feature Set for DDoS Classification Using Machine Learning and Deep Learning
submitted to IEEE 62nd Globecom 2021
-
Towards Network Traffic Monitoring Using Deep Transfer Learning
in 2020 IEEE 19th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom), Guangzhou, China [Acceptance Rate 25%]
-
Smart Home Networks: Security Perspective and ML-based DDoS Detection, Nominated for Best Paper Award
33rd IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), London, ON, Canada, pp. 1-8