{"id":6361,"date":"2023-04-19T09:27:59","date_gmt":"2023-04-19T09:27:59","guid":{"rendered":"https:\/\/www.eucnc.eu\/?page_id=6361"},"modified":"2025-05-23T16:47:42","modified_gmt":"2025-05-23T16:47:42","slug":"special-session-12","status":"publish","type":"page","link":"https:\/\/www.eucnc.eu\/programme\/special-sessions\/special-session-12\/","title":{"rendered":"Special Session 12"},"content":{"rendered":"

Building Blocks for Autonomous Networks<\/h2>\n

Thursday, 5 June 2025, 11:00 \u2013 12:30, room 1.2–1.4<\/h6>\n
Session Chair:<\/h5>\n
    \n
  • Marcin Dryja\u0144ski <\/strong>(Rimedo Labs, PL)<\/li>\n<\/ul>\n

    There is a huge demand for reliable, efficient, and intelligent wireless networks. However, due to the significant increase in their complexity, traditional network management approaches struggle to keep pace with the challenges of contemporary and future networks. In that context, autonomous networks are a highly practical idea that may help in enhancing performance and efficiency in the management of wireless networks. Autonomous wireless networks can self-optimize, self-heal, and self-configure, reducing the need for constant human intervention. Next, by leveraging AI and machine learning to dynamically allocate resources, mitigate interference, and improve spectrum utilization, autonomous networks lead to lower latency and better overall network performance but also can detect anomalies in real-time, proactively identify and neutralize threats, and adapt to new attack vectors without human intervention. Finally, autonomous networks lead to cost reduction by minimizing the need for manual monitoring, maintenance, and troubleshooting. In a nutshell, as connectivity needs grow, the shift toward self-managing, AI-driven wireless networks is inevitable. Autonomous wireless networks are essential to ensure scalability, security, and efficiency in the digital age, paving the way for a smarter and more connected world.<\/p>\n

    Programme<\/h5>\n

    16:00 – 16:30: Keynote talk by S\u0142awomir Sta\u0144czak<\/strong>, prof., TU-Berlin
    \n<\/strong>Title of the talk
    \nFrom Digital Twins to Automated Networks<\/strong>
    \nModern networks are becoming too complex to manage manually. This talk outlines the need for automation, and highlights the digital twin as a key enabler – combining real-time insights with intelligent control. We go beyond digital twins to explore the role of AI in driving autonomy and illustrate the benefits through use cases such as energy efficiency, load balancing, and network slicing. Finally, we show how automation is already transforming networks today.<\/p>\n

    16:30 – 16:45: Talk by Heiko Lehmann<\/strong>, PhD, Deutsche Telekom Group Technology, T-Labs
    \n<\/strong>Title of the talk:
    \nAI for Network Self-Organization<\/strong>
    \nThe notion of autonomous networks relies substantially on self-organization capabilities. As we learn from Statistical Physics, these are closely related to complexity generation phenomena such as the currently much-discussed logical control conflicts. The talk studies these capabilities in modern disaggregation architectures. The role of distributed AI agents is discussed in the sketched \u202fsetting.<\/p>\n

    16:45 – 17:00: Talk by Marcin Dryja\u0144ski<\/strong>, PhD, Rimedo Labs
    \n<\/strong>Title of the talk:
    \nMulti-scale, hierarchical O-RAN rApp-xApp tandem for Energy Saving
    \n<\/strong>In this talk, we discuss the outcome of the T-Labs CIRCAC (Cross-RIC App Collaboration) project which focuses on scale and multi-level network optimization. An O-RAN hierarchical Cell On\/Off Switching (COOS) rApp-xApp tandem operation in an energy-saving use case is presented. The idea behind this xApp\/rApp cooperation is the synergy it brings. While the xApp provides fast, localized control with significant energy savings, the rApp can enhance dynamic adaptability and stability. The COOS-rApp gathers network and user KPIs from a wide network to generate a set of thresholds for cell on and cell off to be provided to COOS-xApp, which in turn decides locally on actual cell on or cell off. Hierarchy is also visible through the decision timing for both Apps: while COOS-rApp operates in a timescale of several minutes, the COOS-xApp can quickly make decisions in the order of seconds or less.<\/p>\n

    17:00 – 17:15: Talk by Martin Stahn<\/strong>, PhD, Deutsche Telekom Group Technology, T-Labs
    \n<\/strong>Title of the talk:
    \nOn the utility of specialized and general online network digital twins<\/strong>
    \nIn future networks, it is expected that many heterogeneous autonomous agents (AI based or else) operate and optimize different objectives of telecommunication networks. This function disaggregation brings a lot of potential for synergies and optimization but also the risk of catastrophic conflicts. On the one hand, specialized online network digital twins can be used to improve the stability and performance of agents focused on a specific subset of objectives, but they might not be sufficient to resolve conflicts between agents targeting vastly different objectives. On the other hand, general online network digital twin could enable more sophisticated conflict resolution approaches, but they are some limitations to overcame to construct sufficient ones.<\/p>\n

    17:15 – 17:30: Talk by Marcin Hoffmann<\/strong>, Rimedo Labs, Poznan University of Technology
    \n<\/strong>Title of the talk:
    \nWhy do we need a unified energy efficiency testing framework for O-RAN hardware and software?<\/strong>
    \nNowadays there is a growing interest in increasing energy efficiency in mobile networks. Especially, the deployment of O-RAN is expected to provide high gains in this field. First, by network hardware virtualization. Second, by deployment of RAN Intelligent Controller (RIC) along with dedicated xApps and rApps, aimed at providing energy savings by adapting network hardware to the network traffic load through, e.g., cell on\/off switching, or RF Channel Reconfiguration. However, the energy efficiency of both apps and hardware components is hard to be compared, because each vendor follows its own scenarios and evaluation procedures. In this talk we will demonstrate the end-to-end energy efficiency testing framework, being a result of the joint work between Rimedo Labs and i14ylab. During the talk, we will discuss the state-of-the-art effort from various standardization bodies like 3GPP, ETSI, and O-RAN ALLIANCE toward energy efficiency testing in 5G\/6G. Based on that we will describe the foundations of the E2E energy-efficiency testing framework, providing the 4 views on this challenge: component-level, global\/feature-base, deployment view, and multi-link\/wide network. The talk aims at starting a broad discussion on how the proposed framework can be extended and adjusted, possibly also to the new network concepts like cell-free networks.<\/p>\n<\/div><\/div><\/div><\/div><\/div>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":2,"featured_media":0,"parent":521,"menu_order":45,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-6361","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/pages\/6361","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/comments?post=6361"}],"version-history":[{"count":19,"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/pages\/6361\/revisions"}],"predecessor-version":[{"id":13190,"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/pages\/6361\/revisions\/13190"}],"up":[{"embeddable":true,"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/pages\/521"}],"wp:attachment":[{"href":"https:\/\/www.eucnc.eu\/wp-json\/wp\/v2\/media?parent=6361"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}