Students will mature a critical view of computer network technologies and performance and acquire the fundamentals of machine learning - applied to network measurements. In particular, students will get: 1. the knowledge of the analysis of Internet configurations and misconfigurations and protocols like IP, TCP, UDP, HTTP, and DNS. 2. the knowledge of performance-related problems in local, wired, and WiFi networks and wide area networks via delay and throughput measurements 3. the ability to process real traffic traces and extract information related to performance and troubleshooting issues from real traffic. 4. the ability to understand and adequately use machine learning tools to solve some classification problems applied to internet traffic analysis. 5. the ability to use modern tools and programming languages like Python and Jupyter Notebook for quick prototyping of high-level machine learning problems. The ability to apply the gained knowledge will be verified by preparing lab reports and discussing it during an oral examination at the end of the course.

Students must know the protocols and mechanisms normally used on the Internet. In particular, students must be experts with LAN protocols (Ethernet IEEE 802.3, WiFi IEEE 802.11, CSMA/CD, direct and indirect delivery, LAN interconnections), network (IPv4, routing protocols), and transport (TCP/UDP) protocols, and application layer protocols (DNS, HTTP, HTTPS, multimedia streaming protocols, etc.). We will also leverage traffic models, and students must be familiar with basic traffic modeling techniques (Poisson models, Markovian models, hypothesis tests, confidence interval evaluation). There are no particular prerequisites for the second part of the course on Machine Learning applications to network management problems.

Students will work in the Laboratory using PCs running Linux OS. Working in groups, students will set up experiments configuring LANs using Ethernet switches and WiFi access points. Each group of students will use 3 PCs (*) in this first part of the lab. Next, students will use real traffic traces to deal with more realistic scenarios. Students will go through the following group of laboratories: • Configuration of hosts in Local Area Networks, IP addresses management, subnetting/supernetting (2 hours) • Traffic monitoring using sniffers in LAN: TCP, UDP and HTTP (6 hours) • Performance measurement for file transfers in wired networks (6 hours) • Performance measurement for file transfers in WiFi setup (8 hours) • Performance measurement for file transfers and impact of delay and packet losses (8 hours) • Analysis of traffic traces collected from real networks and development of post-processing tools to extract information out of the raw data (15 hours) • Usage of statistical methods and data mining techniques to find correlations and solve classification problems (15 hours)

This course comprises a large number of laboratory activities on the topics of communication networks. Students perform these laboratory activities at the LED laboratories of the Electronic and Communication Department. It will also be possible to use your own laptop for laboratories, LED labs, or at home. Before each lab session, the fundamentals of technologies and methodologies will be presented in classes to recap the involved technologies.

The teaching material will be made available by the class teacher on the Didattica web portal. The material will guide the students with a description of the lab experiments, reference documentation, and theoretical information. No textbook is available. Students must be familiar with Internet protocols and applications, which can be found for example, in -- A. Pattavina: Reti di telecomunicazioni, Mc.Graw-Hill (in Italian) -- J.F. Kurose, K.W. Ross: Computer Networking: A Top-Down Approach Featuring the Internet, Pearson (in English) For the Machine Learning part, we suggest following the book -- Wes McKinney: Python for Data Analysis, O'Reilly (in English- available freely online)

Exam: Compulsory oral exam; Individual essay; Group essay;

During the laboratories, students are required to work in groups. At the end of the course, each group will have to deliver a written report on the laboratory experiences indicated by the teacher during the course (group report). In total, the report focuses on eight topics. The report must be delivered in a single PDF file to be uploaded to the teaching portal by the exam registration deadline. The report will be evaluated to verify the correctness of results and the completeness in the description of the performed experiments. All students in the group will have the same mark, expressed in thirtieths, up to a maximum of 27/30. The group report mark will be valid for all group members and will be valid for two years. Once corrected, the report can be changed only if all students agree. Each student will also have to prepare an individual report that must be uploaded as a single PDF file on the teaching portal by the deadline for the student registration for the exam. The individual report will be discussed during the oral examination. During classes and labs, students are invited to anwer questions and discuss the results they obtained, gaining up to 3 points for the final mark . During the oral exam, each student will be asked to discuss the content of the reports (both group and individual). The oral exam is individual, and students will be asked to answer questions about the course's practical and theoretical parts. The grade for the oral exam must be sufficient to pass the exam, with a minimum 18/30 and a maximum 30 cum laude. The final grade will be a weighted average between the evaluations of the group laboratory report (70%) and oral examinations (30%). It is possible to get additional points with specific topic reports or prepare lesson notes to be reused in subsequent years.

In addition to the message sent by the online system, students with disabilities or Specific Learning Disorders (SLD) are invited to directly inform the professor in charge of the course about the special arrangements for the exam that have been agreed with the Special Needs Unit. The professor has to be informed at least one week before the beginning of the examination session in order to provide students with the most suitable arrangements for each specific type of exam.