Detection and Prediction of Network Vulnerabilities with Machine Learning Models and Algorithms

The project investigates the development of artificial intelligence models and algorithms to analyze telecommunication networks, looking for signs that indicate the presence, or imminent arrival, of faults and outages on the network.
The project will use as its main input data (network topology and network health metrics) collected by EXFO in real-time and accumulated over extensive periods of time. Using the collected data, the project will investigate the application of artificial intelligence and, in particular, machine learning techniques to detect/predict network faults, performance anomalies, and even cyber threats, in real-time. Based on the detection and prediction of network faults, we expect to design a cognitive system that will be able to recommend possible mitigations and solutions to overcome those faults, as much as possible. In non-real-time, we plan to design the system in such a way that it will be able to identify network areas with poor system performance and reliability, in order to perform pro-active maintenance.

Faculty Supervisor:

Brigitte Jaumard;Tristan Glatard

Student:

Raj PATEL;Thai Ba PHAM;Hai Hong Vu PHAN;Dat LE;Huy Quang Doung

Partner:

EXFO

Discipline:

Engineering - computer / electrical

Sector:

Information and cultural industries

University:

Concordia University