BONSAI Lab PoliMi

Collaboration with industrial partners remains an important part of the CommTech@PoliMi (Telecommunication Engineering program at Politecnico di Milano) - this collaboration extends across various research projects and courses offered. This Tuesday, Telecommunication Engineering students who attend the Communication Network Design course (taught by Massimo Tornatore) in collaboration with TEA - Telecommunication Engineering Association visited Nokia Vimercate and got the chance to closely see the various laboratories (Optics lab, Microwave lab, PIC lab, Anechoic chamber, and Innovation Hub) conducting R&D in the corresponding telco area. Thanks to Nokia for hosting, and a special thanks to Michael for putting all this together!

Papers can be registered until Nov 12th and submitted by Nov 15th! #ICC25 #IEEE Optical Networking Technical Committee (ONTC)

[People at CommTech@PoliMi series] Meet Francesco Musumeci Francesco is a Professor at Dipartimento di Elettronica, Informazione e Bioingegneria - Politecnico di Milano and shares with us some insights into the course ''Network Data Analysis Lab'' through ten questions. Check out the video to discover how telecom engineers can leverage network data and explore various use cases for applying Artificial Intelligence within telecom networks. --------------------------------------------------------------------------- Francesco è Professore presso il Dipartimento di Elettronica, Informazione e Bioingegneria - Politecnico di Milano e ci condivide alcune informazioni sul corso “Network Data Analysis Lab” attraverso dieci domande. Guarda il video per scoprire come gli ingegneri delle telecomunicazioni possono sfruttare i dati di rete ed esplorare vari casi d’uso dell’applicazione dell’Intelligenza Artificiale nelle reti di telecomunicazione. https://lnkd.in/dMVMF_We

Check out and follow the official page of the Telecommunication Engineering program at Politecnico di Milano. Subscribe to the YouTube channel of CommTech@PoliMi to get stories from professors, researchers, and students. In the video below, Prof. Matteo Cesana, the coordinator of the MSc program, shares his thoughts on what it means to study telecommunications at PoliMi.

BONSAI Lab PoliMi at @ECOC2024 in Frankfurt, Germany We are proud to share with you all that we will be at ECOC-2024 and present six contributions from our team. The list of contributions is as follows: [1] Wednesday, 25-09-2024 at 16.30, W4B session Tutorial on "Machine-Learning-Assisted Optical Network Failure Management: Challenges and Pitfalls," by Francesco Musumeci [2] Wednesday, 25-09-2024 at 14.00, W3E session "Open Implementation of a Large Language Model Pipeline for Automated Configuration of Software-Defined Optical Networks," by Nicola Di Cicco, Mëmëdhe Ibrahimi, Sebastian Troia, Francesco Musumeci, Massimo Tornatore [3] Tuesday, 24-09-2024 at 13.45, Tu3E session "Integrating PPE and Input Refinement for Enhanced QoT Estimation and Optimization in an Optical Mesh Network," by xin yang, Tarek Eldahrawy, Chenyu SUN, Abdel Lorences-Riesgo, Massimo Tornatore, Gabriel Charlet, Yvan Pointurier [4] Wednesday, 25-09-2024 at 14.00, Poster session P101 "On High-Power Optical Amplification in Hollow Core Fibers for Energy Efficiency and Network Throughput Maximization," by Giovanni Sticca, Mëmëdhe Ibrahimi, Nicola Di Cicco, Francesco Musumeci, Massimo Tornatore [5] Wednesday, 25-09-2024 at 14.00, Poster session P77 "Investigation of SOA Nonlinear Impairments Dependency on Transmission System Parameters," by Xiaohui Zhao, Maha Bouhadida, Hartmut Hafermann, Loig Godard, Iosif Demirtzioglou, Abel Lorences-Riesgo, Zhenzhen Zhang, Massimo Tornatore, Yann Frignac, Gabriel Charlet [6] Wednesday, 25-09-2024 at 14.00, Poster session P169 "Shared-Protected Backup Paths Assignment with Mode Group Division Multiplexing in Optical Networks," by Jiaheng Xiong, Qiaolun Zhang, Ruikun Wang, Alberto Gatto, Francesco Musumeci, Massimo Tornatore See you next week!

A great honor to be part of this project reshaping the future of programmable networks.

The first open-source dataset on hardware failures in microwave networks!!! https://lnkd.in/dki7gSme We are proud to share with you our latest open-access article, "Machine Learning for Failure Management in Microwave Networks: a Data-Centric Approach," published in IEEE TNSM. This work is a product of an ongoing collaboration between Dipartimento di Elettronica, Informazione e Bioingegneria at Politecnico di Milano and SIAE MICROELETTRONICA, on the application of Machine Learning in microwave networks. 1️⃣ We make a case for Data-Centric Machine Learning for failure management in microwave networks, and we focus on hardware failures (vs. mostly propagation failures in past work). 2️⃣ We leverage the predictive uncertainty for quantifying the information gained by labeling batches of data points via uncertainty-based Active Learning. Our approach allows training a top-performing model requiring only ∼15% of the full dataset, thus granting proportional savings in annotation times and costs. 3️⃣ We leverage Conditional Variational Autoencoders for generating synthetic data to enrich poorly represented failure classes, with the goal of improving the average classification performance among all failure classes. 4️⃣ We make our high-quality dataset (curated by microwave industry experts) publicly available. 5️⃣ We remain committed to promoting open science by making our publication open-access for everyone to read. You can find the complete article in the following link: https://lnkd.in/daH9maft

OFC Conference is in a week, and we are ready to go to San Diego! We are proud to announce the schedule of the seven accepted papers at OFC 2024! Here's the schedule and paper details: [1] 26 March at 17.00. Tu3B: 6G and Emerging Applications "Hollow-Core-Fiber Placement in Latency-Constrained Metro Networks with edgeDCs" by Giovanni Sticca, Mëmëdhe Ibrahimi, Nicola Di Cicco, Francesco Musumeci, and Massimo Tornatore [2] 26 March at 17.15. Tu3B: 6G and Emerging Applications "Network for AI: Communication-Efficient Federated Learning with MST-based Scheduling and Multi-Aggregation over Optical Networks" by Ruikun Wang, Jiawei Zhang, Mëmëdhe Ibrahimi, Zhiqun Gu, Yuming Xiao, Francesco Musumeci, Massimo Tornatore, and Yuefeng Ji [3] 26 March at 18.00. Tu3B: 6G and Emerging Applications "Availability-Guaranteed Differentiated Provisioning in Integrated Satellite-Terrestrial Optical Networks" by Lu Zhang, Xin Li, Massimo Tornatore, Jingjie Xin, and Shanguo Huang [4] 27 March at 17.30. W4I: AI-Based Automation "Digital Twin-based Insertion Loss Estimator for Anomalous Loss Localization and Network Equalization Enhancement" by xin yang, Chenyu Sun, Gabriel Charlet, Massimo Tornatore, and Yvan Pointurier [5] 27 March at 18.15. W4I: AI-Based Automation "Resource Re-Allocation for Pre-Planned Power Outages in Optical Networks" by Qiaolun Zhang, Patricia Layec, Achille Pattavina, and Massimo Tornatore [6] 28 March at 08.45. Th1I: Next Generation ROADMs, Multiband and SDM Networking "Throughput Increase in Multi-Fiber Networks using Partial Lane-Change Capabilities" by Oleg Karandin, Francesco Musumeci, Yvan Pointurier, and Massimo Tornatore [7] 28 March at 15.30. Th3I: Survivability and Fault Management "DC-Carrier Cooperation for Rapid Restoration against PNE-Node Failure in Optical Networks" by Subhadeep Sahoo, Sifat Ferdousi, Sugang Xu, Yusuke Hirota, Massimo Tornatore, Yoshinari Awaji, and Biswanath (Bis) Mukherjee

🌿 BONSAI Research Group at NOKIA Italy: AI for Network Management 🌿 On February 8th, Massimo Tornatore and Francesco Musumeci led an engaging lecture titled "AI for Network Management" at Nokia Italy. The lecture delved into: **The Fundamental Concepts of Machine Learning (ML)** - laying the groundwork for understanding its impact **ML Applications in Optical Networks** - showcasing innovations that are driving efficiency and enhancing optical network operations **ML Applications in Microwave Networks** - exploring cutting-edge solutions for network optimization and showcasing the tangible benefits of our research endeavors **A Broad Overview of Other Applications** - highlighting the versatility and potential of ML across various aspects of network management We extend our heartfelt thanks to Nokia for hosting us and everyone who joined us. #AIfornetworkmanagement #BONSAIResearchGroup #NokiaItaly #MachineLearning #OpticalNetworks #MicrowaveNetworks #CommTechPoliMi

GASTON: Green, Autonomic, and Resilient Optical Multi-Core Networks We are pleased to announce the successful kick-off meeting of the GASTON project on Wednesday (14-02-2024) at Politecnico di Milano, Milano. At the kick-off meeting, we were pleased to host Andrea Marotta, Cristian Antonelli, and Giammarco Di Sciullo from Università degli Studi dell'Aquila. The GASTON project is a collaboration between Politecnico di Milano and Università degli Studi dell'Aquila, led by Massimo Tornatore and Andrea Marotta. The main Challenges Addressed by GASTON are: 1️⃣ Guaranteeing environmental sustainability through innovative infrastructure resilience strategies. 2️⃣ Ensuring stakeholder cooperation via privacy-preserving approaches to achieve high-capacity, energy-efficient, and survivable Multi-Core Optical Networks. Stay tuned for more news on this research activity toward greener and more resilient multi-core optical networks! #GASTONProject #OpticalNetworks #BONSAIResearchGroup #PoliMi #UnivAq #CommTechPoliMi Massimo Tornatore, Andrea Marotta, Cristian Antonelli, Francesco Musumeci, Mëmëdhe Ibrahimi, Giammarco Di Sciullo, Giovanni Sticca, Oleg Karandin