Description
Objectives
Develop competences on collecting, analysis, modeling and management of traffic in data networks. Learn to use advanced statistical methods to identify and classify network data traffic and anomaly detection. Learn technologies of traffic control and management widely used by teleccommunication operators.
Syllabus
1 - Traffic models: Poisson models, MMPP, ON-OFF, auto similarity, MMPP, Modelos de tráfego: modelos de Poisson, ON-OFF, MMPP, auto-similar. 2 - Routing in MPLS. VPN-MPLS. Traffic engineering with MPLS.Notion of segment routing. 3 - Introduction to programmable networks. Notion of network operating system and centralized control in networks. Topics on network function virtualization. 4 - Statistical and machine learning methods for traffic analysis: regression and classification. Supervised learning: KNN, naive bayes, neural networks. Unsupervised learning. Clustering techniques: k-means. Notion of generalization and regularization. Application to flow identification and intrusion detection.
Prerequisites
Students must have previousy completed the curricular unit of Data Networking and Internet and must have good programming background.
Cross Competence Component
The UC contributes to the development of transversal competences mainly on intrapersonal skills (team work, oral presentation) and Information and Media Literacy (ability to locate and access information, as well as to analyze and evaluate media content) within the lab projects. Critical thinking will also be required and leveraged. The assessment percentage associated with these competences should be around 10%.
Laboratorial Component
4 small projects, each one covering one subject topic. Each project will take bewteen 4 and 8 hours of lab classes, plus autonomous work.
Programming And Computing Component
Part of the lab projects will make use of either high level scientific languages (Matlab, Octave, R) or general purpose languages (python, java, C) depending on the specific lab project. The programming component will take about 30% of each lab project, while the remaining 70% will be filled by system design and configuration/parametrization.
Ethical Principles
All members of a group are responsible for the group’s work In any assessment, every student shall honestly disclose any help received and sources used. In an oral assessment, every student shall be able to present and answer questions about the entire assignment and solution.