The class of Projects and Laboratory on Photonic Networks, PLPN in the following, will introduce the concepts of physical-layer-aware networking needed to maximize the capacity of back-bone and extended-metro fiber optics networks. PLPN aims at describing peculiarities of data networking based on the exploitation of photonic transmission on optical fiber networks. With the specific purpose of multilayer optimization down from the IP layer, enabling full exploitation of the photonic layer either using the state-of-the art WDM fixed-grid, either the already standardized flex-grid. The concept of Software Defined Network will be introcuded together with the concept of line-system controller. The teaching method will follow an application-oriented introduction of concepts. To this purpose, students will be required to apply the acuired knowledges by developing Phyton modules performing simple network control operations. These will be the homeworks used to the student assessment. Coding will be addressed to the standard open-source procedure based on GitHub. Lectures on Python coding and use of Github will be part of the class. Seminars will be given by companies and operators operating in the field. In particular, by Facebook, Cisco, SMOptics, Coriant Networks, OpenFiber and TIM. The final student assessment will be done through the discussion on the assigned homework. For PLPN students will be available a set of homeworks that may evolve into a Master thesis work, being its initial phase. The PLPN class will take advantage of the experience gained participating to the consortium Telecom Infra Project.
o State-of-the art transceivers for optical communications
o Foundations of optical fiber propagation and modeling its impairments
o Amplifiers and passive components
o WDM spectral use and standards
o ROADMs and node structure in general
o YANG, Netconfig, GMPLS, OTN
o SDN controller and line-system controller
o Python language
o TensorFlow libraries for machine learning
o Python development within GitHub
o Emulation of optical layer in photonic networks
o Routing spectral and wavelength assignment
o Multilayer optimization, including machine learning methods.
o In general, ability to perform physical-layer-aware network analysis, design and optimization
This class will need foundation of signal analysis and digital transmission as well as general knowledge of the Internet structure. If selected students will miss some of the prerequisites, specific summary session on selected topics will be organized.
Teaching method will be “hands-on”, so within every lecture students will be required to use their own laptop so that theoretical concept will be immediately applied in simple exercises or reviewing examples. For approximately 1/3 of the available hours, the main teacher will be asssisted by assistants supporting code development and in general exercise solving. In the following, the course syllabus as a list of arguments is shown.
o Photonic networks: classification and structure.
o Introduction to the SDN paradigm
o Passive and active network elements
o Fundation of fiber propgation: close form models
o Fundation of Python coding
o The Qaulity-of-Tranmission estimator for lightpaths (related homework)
o The concept of line-sytem controller (related homework)
o The Statistical Network Assessment Process
o Telemtetry and machine learning (related homework)
o Protocols and network layers in photonic networks
o Open-source and networking: the Telecom Infraproject of Facebook
o Companies' seminars (Possible companies are: Cisco, SMOptics, Juniper, Coriant, TIM, OpenFiber, Orange, Microsoft, Facebook)
The class will be organized as a series of concepts’ presentation and their application through Python coding homework.
Students will be required to operate on their own laptop and group working will be allowed.
Teaching material will be available on “portale della didattica”. Books and websites to deepen specific topics will be suggested as well.
Modalità di esame: prova orale obbligatoria; progetto individuale; progetto di gruppo;
Exam: compulsory oral exam; individual project; group project;
Student assessment will be performed reviewing the Python coding homework, including the proper use of github and reports. This work can be a group work. This process will deliver a maximum of 25 points. The remaining 5 points – and possible laude – will be assigned during the individual final oral discussion.