NVS2 – 6G Network Technology, Architecture & Infrastructure
Wednesday, 3 June 2026, 17:00-18:30, room Sala 4 (1st floor)
Session Chair: Marja Matinmikko-Blue (University of Oulu, Centre for Wireless Communications, FI)
AI Sessions for Network-Exposed AI-as-a-Service
Merve Saimler (Ericsson Research, Turkey); Mohaned Chraiti (Sabanci University, Turkey)
Cloud-based Artificial Intelligence (AI) inference is increasingly latency- and context-sensitive, yet today’s AI-as-a-Service is typically consumed as an application-chosen endpoint, leaving the network to provide only best-effort transport. This decoupling prevents enforceable tail-latency guarantees, compute-aware admission control, and continuity under mobility. This paper proposes Network-Exposed AI-as-a-Service (NE-AIaaS) built around a new service primitive: the AI Session (AIS)-a contractual object that binds model identity, execution placement, transport Quality-of-Service (QoS), and consent/charging scope into a single lifecycle with explicit failure semantics. We introduce the AI Service Profile (ASP), a compact contract that expresses task modality and measurable service objectives (e.g., time-to-first-response/token, p99 latency, success probability) alongside privacy and mobility constraints. On this basis, we specify protocol-grade procedures for (i) DISCOVER (model/site discovery), (ii) AI PAGING (context-aware selection of execution anchor), (iii) two-phase PREPARE/COMMIT that atomically co-reserves compute and QoS resources, and (iv) make-before-break MIGRATION for session continuity. The design is standard-mappable to Common API Framework (CAPIF) style northbound exposure, ETSI Multi-access Edge Computing (MEC) execution substrates, 5G QoS flows for transport enforcement, and Network Data Analytics Function (NWDAF) style analytics for closed-loop paging/migration triggers.
Leveraging on-Site EVs for Enhanced Stationary Battery Utilization and Reliability in Base Stations
Mohammad Reza Jokar, Greta Vallero, Daniela Renga and Michela Meo (Politecnico di Torino, Italy)
High-reliability telecom base stations (BSs) are a key enabler of sustainable 6G networks, yet guaranteeing resilience to power grid outages often forces their stationary battery energy storage systems (BESSs) to operate conservatively, leading to underutilization during normal operation. This paper presents a novel framework that integrates electric vehicles (EVs) parked at BS sites as a dynamic energy reserve, improving BS resilience during power grid outages while simultaneously reducing operating costs under normal conditions. By leveraging EV availability to withstand prolonged outages, the stationary BESS can safely operate at deeper depths of discharge (DOD) without compromising backup reliability, enabling energy arbitrage when the grid is available. The proposed methodology incorporates detailed models of battery aging, stochastic power grid outages, and probabilistic EV availability. Comparative case studies show that EV involvement reduces unmet energy demand during outages by 74% and more than triples the served traffic, while lowering total operating costs by 6.7% during normal operation. These results demonstrate that EV-assisted BESS management effectively decouples resilience requirements from economic operations, thereby strengthening the robustness and cost efficiency of next-generation telecommunications infrastructure.
Energy Modeling of a Self-Sustainable Base Station
Juho Markkula, Sanna Tuomela, Hamid Malik and Antti Pauanne (University of Oulu, Finland)
Electrical grid may not be available for providing power for cellular networks in remote locations such as forests and mountains. However, having connectivity in this kind of wilderness areas would be beneficial, for example, for being able to connect to the Internet and having voice calls. Alternatively, renewable energy sources such as wind, solar, and hydrogen could be utilized to power up the base station. This paper studies with Matlab simulations the energy modeling of a self-sustainable base station. Two different selections of equipment for renewable energy sources to provide sufficient energy for an open radio access network (O-RAN) base station for the whole year with extremely varying weather conditions in Oulu, Northern Finland were evaluated.
Advances on Unified Architecture for Open RAN-Enabled Distributed and Scalable 6G Networks
Luis Blanco (Centre Tecnològic de les Telecommunicacions de Catalunya (CTTC), Spain); Cristian J. Vaca-Rubio (Centre Tecnologic de Telecomunicacions de Catalunya (CTTC), Spain); Jorge Baranda (Centre Tecnològic de Telecomunicacions de Catalunya (CTTC/CERCA), Spain); Farhana Javed (Centre Tecnològic de Telecomunicacions de Catalunya (CTTC), Spain); Selva Via and Engin Zeydan (CTTC, Spain); Luis Contreras (Telefonica, Spain); Javier Velázquez Martínez (Telefonica Innovacion Digital, Spain); Gil Kedar (Ceragon, Israel); Efi Dvir (Ceragon Networks, Romania); Gianluca Rizzo (HES SO Valais, Switzerland & Universita’ di Torino, Italy); Berk Buzcu (HES-SO Valais-Wallis, Switzerland); Maria A. Serrano, Angelos Antonopoulos and Godfrey Mirondo Kibalya (Nearby Computing, Spain); Jetmir Haxhibeqiri, Vasilis Maglogiannis and Dries Naudts (IMEC, IDLab, Ghent University, Belgium); Jeroen Hoebeke (Ghent University – imec, Belgium); Tomaso De Cola and Balaji Kirubakaran (German Aerospace Center (DLR), Germany); Sotirios Spantideas (Four Dot Infinity, Athens, Greece); Anastasios E. Giannopoulos (Four Dot Infinity, Greece); Panagiotis Trakadas (Four Dot Infinity, Athens, Greece); Harilaos Koumaras (NCSR Demokritos, Greece); Stefanos Plastras (National Center for Scientific Research Demokritos, Greece); Alice Piemonti (Martel Innovate, Switzerland); Vito Cianchini (Martel Innovate Gmbh, Switzerland); Carla Amatetti, Bruno De Filippo and Zefu Gao (University of Bologna, Italy); Pawel Kryszkiewicz, Marcin Pakula and Łukasz Kułacz (Poznan University of Technology, Poland); Paweł Płaczkiewicz (Poznań University of Technology, Poland); Adrian Kliks (Poznan University of Technology, Poland); Matteo Pagin and Hua Wang (Keysight Technologies, Denmark); Sihem Cherrared and Taki Djaidja (Orange Labs France, France)
UNITY-6G introduces a AI-natively framework that unifies terrestrial (TN), non-terrestrial (NTN), and non-public networks (NPN), treating connectivity, computing, and intelligence as interdependent resources. The architecture utilizes an Inter-Domain Management Orchestrator (IDMO) based on Service-Based Management Architecture (SBMA) principles to coordinate services across heterogeneous domains. A core pillar of the framework is its AI-native design through autonomous agentic AI workflows following a standardized MS-AE-DE-ACT (Monitoring, Analytics, Decision, and Actuation) logical patterns. To enhance resource efficiency and sustainability, the architecture integrates Digital Twins (DT) for proactive system modeling and semantic communications to prioritize task-relevant information transfer. Security is addressed through a Trust Architecture leveraging Distributed Ledger Technology (DLT) for cross-domain auditability. The framework’s utility is validated through proof-of-concepts targeting sustainable disaster handling, immersive XR/holographic communications, and time-sensitive services for Industry 4.0. The presented advances establish a foundation for the continuous development of high-performance, autonomous 6G systems.
Evolution in Security and Privacy for 6G Networks: AI Perspectives and Key Enablers
Charuka Moremada and Vidura Ravihansa (University College Dublin, Ireland); Bartlomiej Siniarski (MBP Network Technology, Ireland); An Braeken (Vrije Universiteit Brussel, Belgium); Viet Quoc Pham (Trinity College Dublin, Ireland); Yushan Siriwardhana (University of Oulu, Finland); Basak Ozan Ozparlak (Ozyegin University, Turkey); Eirini Kanaki (Z-RED & Tech4Society, Greece); Gürkan Gür (Zurich University of Applied Sciences (ZHAW), Switzerland); Periklis Chatzimisios (International Hellenic University, Greece & University of New Mexico, USA); Edgardo Montes de Oca (Montimage EURL, France); Selim Sarfati (Ozyegin University, Turkey); Madhusanka Liyanage (University College Dublin, Ireland)
Artificial Intelligence (AI) is a key enabler of intelligent management, optimization, and security in the evolution of sixth-generation (6G) networks. Although AI can significantly enhance security and privacy, it also introduces new vulnerabilities and challenges related to sustainability, regulation, and ethics. This paper presents an interdisciplinary review of AI-driven security and privacy mechanisms for 6G, and analyzes emerging threats, countermeasures, and architectural enablers for secure and privacy-aware services. It further discusses the sustainability and legal implications of AI-based solutions and identifies key research directions for trustworthy AI-powered 6G ecosystems. Going beyond a state-of-the-art (SoTA) survey, this paper includes an experimental feasibility analysis of Large Language Model (LLM)-based intrusion detection for future networks, aligned with recent advances in Generative AI (GAI).
