Design and improvements of security solutions to be integrated in an Internet gateway

Four research projects addressing issues related to computer network and web security are proposed. In Project 1, the development of a novel self-learning anomaly detection tool is proposed to protect web servers against emerging elaborate attacks (such as spamming, multi-step and zero-day attacks). This tool will be developed using Machine Learning and Data Mining techniques. In Project 2, an event correlation system for cloud-based infrastructures will be developed. This tool will correlate across several heterogeneous types of cloud, sensors and logs to detect complex attacks using the traces attacks left across systems they compromised. In Project 3, a tool that can automatically analyze and identify malicious web content before it deploys in the target network will be developed using a comprehensive context-aware and ontology-based event correlation approach. In Project 4, we will test a novel malware detection approach based on physical measures (e.g. CPU, memory, network bandwidth, etc.). Groupe Access is looking to use results and findings of these projects to improve the security level of systems it manages. Tools developed will enable early threat detection, improve analysis of ongoing events and display warnings related to abnormal activities.

Faculty Supervisor:

Andrea Schiffauerova;Jean-Marc Robert;José Fernandez

Student:

Partner:

Groupe Access

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

Concordia University; École Polytechnique de Montréal

Program: