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

The project investigates the research and development of static and dynamic autonomous network management agents that will be able to analyze telecommunication network behaviors, and predict faults and outages. Using collected datasets from EXFO network monitoring systems, we will research, analyze, design and develop machine learning models and algorithms adapted to the analysis of time series. These algorithms can be used as part of network monitoring toolsets to dynamically monitor the network behaviors. They will also allow real-time network faults detection/prediction, performance anomalies, and cyber threats. Machine learning models and algorithms will be tailored to assess the potential service impact of these faults, anomalies, and threats. This assessment will lead to the recommendation of mitigations and solutions, along with their expected impact on the network. In non-real-time contexts, we will analyze performance data collected over significant time periods and correlate them with the evolution of the network in terms of variables, such as topology and configuration. This performance data analysis will allow the identification of network areas with poor system performance and reliability.

Design of Rotating Shift Schedules while Mitigating Circardian Cycle Disruption

The main goal of this project is to design rotating shift schedules that mitigate the circadian cycle disruptions, while taking account that the circadian cycle varies from person to person
Project in done in collaboration with Dr. Tristan Glatard (Concordia University) and Dr. Diane Boivin (Department of Medecine, McGill University).

Defragmentation in WDM Optical Networks

As lightpath services are added to a network over time, spectrum becomes fragmented, i.e., available wavelengths on different links do not match and thus no end-to-end paths can be established. While an offline provisioning where all traffic is known can minimize this effect to some extent, a dynamic or even incremental scenario is naturally more prone to creating a fragmented spectrum with wasted capacity.
There are multiple flavours of network defragmentation, depending on the specific deployment and technological constraints. Most commonly, lightpath reconfiguration is considered, wherein existing demands’ spectrum and paths are rearranged to accommodate a known set of additional demands. Several techniques can be employed with this goal: traditionally, a “make-before-break” technique is considered, where the rearranged demand is setup before the original one is torn down in order to prevent traffic disruption, by, e.g., duplicating the existing traffic.