6GV & NET1 – 6G vision building and network transformation & AI powered 6G networks
Thursday, 9 June 2022, 10:30-12:00, Room A304
Session Chair: Seppo Yrjölä (Nokia & Centre for Wireless Communications, University of Oulu, FI), Volker Ziegler (Nokia Bell Labs & CTO, DE)
Visions for 6G Futures: A Causal Layered Analysis
Seppo Yrjölä (Nokia & Centre for Wireless Communications, University of Oulu, Finland); Petri Ahokangas (University of Oulu, Finland); Marja Matinmikko-Blue (University of Oulu, Centre for Wireless Communications, Finland)
This study extends and deepens the joint 6G vision building stemming from use cases, enabling technologies, key performance indicators (KPIs), key value indicators (KVIs) and business scenario litanies towards the social, the worldview and the metaphors layers utilizing causal layered analysis (CLA) method. 6G visions are explored from different national perspectives assessing future initiatives from China, Europe, Japan, South Korea, and US. The multiple ideologies and epistemes of the stakeholders are mapped to create a transformed future vision emphasizing the importance of 6G design from the triple bottom line of sustainability including social, economic and environmental perspectives. Collaborative research, harmonized standardization, and anticipatory regulation efforts were found essential in developing trustworthy and general purpose 6G technologies for users and developers.
Organic 6G Networks Decomplexification of Software-Based Core Networks
Marius Corici and Eric Troudt (Fraunhofer FOKUS, Germany); Thomas Magedanz (Fraunhofer Institute FOKUS / TU Berlin, Germany); Hans Schotten (DFKI, Germany)
This article proposes a new service architecture for 6G core networks based on the advancements in software services adopted by the IT industry. Through reconsidering the 5G Service Based Architecture (SBA) functional split, we propose a new “Organic 6G Network” concept. It provides a fresh perspective on how to achieve the functional goals of 6G networks, while at the same time providing more adaptation capabilities and decreasing the overall system complexity. Specifically, this article concentrates on the complexity analysis, by providing for it a comprehensive concept definition, an understanding of the elements generating it in the current 5G architecture, and by providing a new direction on how to tackle its reduction. A thorough analysis of the new concept is also presented, clearly underlining how the proposed “Organic 6G Network” architecture has the potential to greatly reduce the complexity while at the same time to outperform the flexibility of the current 5G network, making it a high potential alternative for the 6G networks.
Dynamic Network (Re-)configuration Across Time, Scope, and Structure
Roland Bless (Karlsruhe Institute of Technology (KIT), Germany); Bastian Bloessl (TU Darmstadt, Germany); Matthias Hollick (Technische Universität Darmstadt & Secure Mobile Networking Lab, Germany); Marius Corici (Fraunhofer FOKUS, Germany); Holger Karl (Hasso Plattner Institute & University of Potsdam, Germany); Dennis Krummacker (German Research Center for Artificial Intelligence (DFKI GmbH), Germany); Daniel Lindenschmitt (Technische Universität Kaiserslautern, Germany); Hans Schotten (DFKI, Germany); Lara Wimmer (IHP Leibniz Institut fur Innovative Mikroelektronik, Germany)
Today’s wired and wireless networks transport data and delivery services over a wide range of locations, devices types, and usage scenarios. While they do adapt to changing situations, they are, at their core, essentially still static. In this paper, we attempt to collect challenges and opportunities to substantially broaden the adaptation options for networks with respect to both data transport and service delivery. We identify three key dimensions for adaption: time, scope, and structure. We discuss how adaption techniques can deal with foreseeable changes in networks and work along these three dimensions. We also provide speculations on promising solution approaches.
Federated Learning Based Anomaly Detection as an Enabler for Securing Network and Service Management Automation in Beyond 5G Networks
Suwani Jayasinghe, Yushan Siriwardhana and Pawani Porambage (University of Oulu, Finland); Madhusanka Liyanage (University College Dublin, Ireland & University of Oulu, Finland); Mika E Ylianttila (University of Oulu, Finland)
Network automation is a necessity in order to meet the unprecedented demand in the future networks and zero touch network architecture is proposed to cater such requirements. Closed-loop and artificial intelligence are key enablers in this proposed architecture in critical elements such as security. Apart from the arising privacy concerns, machine learning models can also face resource limitations. Federated learning is a machine learning based techniques which address both privacy and communication efficiency issues. Therefore, we propose a federated learning based model incorporating ZSM architecture for network automation. The paper also contains the simulations and its results of the proposed multi-stage federated learning based which use UNSW-NB15 Dataset.
Robust and Resilient Federated Learning for Securing Future Networks
Yushan Siriwardhana and Pawani Porambage (University of Oulu, Finland); Madhusanka Liyanage (University College Dublin, Ireland & University of Oulu, Finland); Mika E Ylianttila (University of Oulu, Finland)
Machine Learning (ML) and Artificial Intelligence (AI) techniques are widely adopted in the telecommunication industry, especially to automate beyond 5G networks. Federated Learning (FL) recently emerged as a distributed ML approach that enables localized model training to keep data decentralized to ensure data privacy. In this paper, we identify the applicability of FL for securing future networks and its limitations due to the vulnerability to poisoning attacks. First, we investigate the shortcomings of state-of-the-art security algorithms for FL and perform an attack to circumvent FoolsGold algorithm, which is known as one of the most promising defense techniques currently available. The attack is launched with the addition of intelligent noise at the poisonous model updates. Then we propose a more sophisticated defense strategy, a threshold-based clustering mechanism to complement FoolsGold. Moreover, we provide a comprehensive analysis of the impact of the attack scenario and the performance of the defense mechanism.