Cybersecurity

In the ever-evolving landscape of cybersecurity threats, researchers at MUET are collaborating across disciplines to develop innovative solutions that protect digital infrastructures and enhance network security. MUET is addressing critical challenges in areas such as intrusion detection, network security, and machine learning applications.

Explainable Intrusion Detection in VANETs

Dr. Fayyaz Mangrio and Dr. Zafi Sherhan Shah from the Telecommunication Engineering Department are leading an interdisciplinary project titled "Achieving model explainability for intrusion detection in VANETs with LIME". This research combines expertise in computer science, telecommunications engineering, and machine learning to improve the transparency and interpretability of intrusion detection systems deployed in Vehicular Ad-Hoc Networks (VANETs).

By leveraging Local Interpretable Model-Agnostic Explanations (LIME), a technique for explaining black-box machine learning models, the project aims to enhance the understanding of how intrusion detection models make decisions and identify potential vulnerabilities or false positives. This interdisciplinary approach contributes to the development of more robust and trustworthy cybersecurity solutions for the rapidly evolving field of connected vehicles.

Real Traffic Dataset for Intrusion Detection

Another interdisciplinary effort, led by Dr. Fahim Yar Khuhawar from the Telecommunication Engineering Department, focuses on the "Development and Validation of Dataset for Intrusion Detection System over Real Traffic." This project involves creating and validating a comprehensive dataset specifically designed for training and evaluating intrusion detection systems (IDS) in real network traffic scenarios.

By combining expertise in computer science, telecommunications engineering, and data analysis, the research team aims to provide a representative dataset that accurately reflects the characteristics of real-world network traffic. This interdisciplinary approach enables researchers and practitioners to develop and assess IDS solutions effectively, contributing to the overall enhancement of network security.

Machine Learning for DHCP DoS Attack Detection

Dr. Khuhawar is also leading a project titled "Machine Learning Approach for Classification of DHCP DoS Attacks in NIDS," which focuses on developing a machine learning-based approach for classifying DHCP (Dynamic Host Configuration Protocol) Denial-of-Service (DoS) attacks within Network Intrusion Detection Systems (NIDS).