Wednesday, 4 June 2025, 12:30-13:00, level 1
Session Chair: Pawel Kryszkiewicz (Poznan Univ. of Technology, PL)
P-1.1 6G NeXt – Split Computing for Smart Drones
Sergiy Melnyk (German Research Center for Artificial Intelligence, Germany); Qiuheng Zhou (German Research Center for Artificial Intelligence (DFKI GmbH), Germany); Hans Dieter Schotten (Deutsches Forschungszentrum für Künstliche Intelligenz GmbH, Germany); Tobias Pfandzelter (Technische Universität Berlin & Einstein Center Digital Future, Germany); David Bermbach (TU Berlin, Germany); Robert Vilter (University of Applied Sciences Wildau, Germany); Nick Stuckert (Technische Hochschule, Germany)
With the rising amount of UAV missions for logistics, agriculture, and other domains, control and anti-collision applications will gain importance. In this paper, we propose a UAV coordination system following a split computing paradigm. Furthermore, a proof-of-concept setup was built, showing promising results for future utilisation of the proposed system.
P-1.2 Efficient Federated Learning Tiny Language Models for Mobile Network Feature Prediction
Daniel Becking (Fraunhofer Heinrich Hertz Institute (HHI), Germany & Technische Universität Berlin, Germany); Ingo Friese (Deutsche Telekom Laboratories, Germany); Karsten Müller (Fraunhofer HHI, Germany); Thomas Buchholz and Mandy Galkow-Schneider (Deutsche Telekom AG, Germany); Wojciech Samek (Fraunhofer Heinrich Hertz Institute, Germany); Detlev Marpe (Fraunhofer Institute for Telecommunications – Heinrich Hertz Institute, Germany)
In telecommunications, Autonomous Networks (ANs) automatically adjust configurations based on specific requirements (e.g., bandwidth) and available resources. These networks rely on continuous monitoring and intelligent mechanisms for self-optimization, self-repair, and self-protection, nowadays enhanced by Neural Networks (NNs) to enable predictive modeling and pattern recognition. Here, Federated Learning (FL) allows multiple AN cells – each equipped with NNs – to collaboratively train models while preserving data privacy. However, FL requires frequent transmission of large neural data and thus an efficient, standardized compression strategy for reliable communication. To address this, we investigate NNCodec, a Fraunhofer implementation of the ISO/IEC Neural Network Coding (NNC) standard, within a novel FL framework that integrates tiny language models (TLMs) for various mobile network feature prediction (e.g., ping, SNR or band frequency). Our experimental results on the Berlin V2X dataset demonstrate that NNCodec achieves transparent compression (i.e., negligible performance loss) while reducing communication overhead to below 1%, showing the effectiveness of combining NNC with FL in collaboratively learned autonomous mobile networks.
P-1.3 Hybrid Multiple Access Scheme Employing MU-MIMO THP/LP, NOMA-MRT and OMA-MRT in Time Selective Fading Channels
Yuka Kimura, Nozomi Sasaki and Hirofumi Suganuma (Waseda University, Japan); Keita Kuriyama (NTT Corporation, Japan); Fumiaki Maehara (Waseda University, Japan)
This study proposes a hybrid multiple access scheme designed to optimize system performance in time selective fading channels by adaptively selecting among three wireless communication methods: multi-user multiple-input multiple-output (MU-MIMO) with linear precoding (LP) and Tomlinson-Harashima precoding (THP), orthogonal multiple access (OMA) and non-orthogonal multiple access (NOMA) which leverage maximum ratio transmission (MRT). The proposed scheme focuses on the difference, in the performance degradation due to time selective fading, between multiple access schemes and dynamically selects the appropriate scheme based on the theoretical system capacity derived from terminal speed information and channel state information (CSI). To validate the proposed system’s effectiveness, we compared its throughput to those of traditional MU-MIMO THP/LP, NOMA-MRT, and OMA-MRT.
P-1.4 RIC Testing as a Platform: Orchestrating a Digital Twin Framework
Mina Yonan, Principal (Principal R&D, Egypt & Orange, France); Kamil Kociszewski (Orange & Orange Innovation Poland, Poland); Aya Kamal and Abdelrhman Soliman (‘5G Software Engineer’, Egypt); Bartosz Rak (Orange Poland, Poland)
RIC Testing as a Platform (RIC-TaaP) has been introduced to tackle the current challenges towards the adoption of RAN Intelligent Controllers (RICs), aiming to elevate their effectiveness and empower Independent Software Vendors (ISVs). RIC-TaaP introduced the standardized closed-loop control interface, diverse testing environment, intelligence and proven Digital Twin (DT) scenarios and test cases. In this poster, we present three examples of calibration with operational networks: SU-MIMO, Ray Tracing, and V2X Handover (HO) Management, along with the automated DT framework within RIC-TaaP. This work highlights the capabilities of the RIC-TaaP platform within the RIC ecosystem showing its potential to foster the RIC adoption and its efficiency.
P-1.5 Hybrid Transmission Scheme Employing OTFS and SC-FDE for Seamless Mobile Communications
Tomohiro Matsuyama, Shogo Kurimoto, Hirofumi Suganuma and Fumiaki Maehara (Waseda University, Japan)
This paper presents a hybrid transmission scheme employing orthogonal time frequency space (OTFS) and single-carrier frequency-domain equalization (SC-FDE) to ensure seamless and reliable communication across diverse terminal mobility conditions and multipath delay profiles. The proposed scheme dynamically switches between OTFS and SC-FDE based on the Doppler frequency. In the OTFS mode, the number of symbols in the delay domain is adjusted according to the maximum multipath delay time, exploiting frequency and time diversity while mitigating inter-symbol interference (ISI). The effectiveness of the proposed approach is validated through computer simulations, demonstrating its superior performance in comparison to OTFS-only and SC-FDE-only scenarios. Thus, the proposed hybrid OTFS/SC-FDE scheme enhances communication reliability across diverse mobility conditions and supports the seamless integration of emerging wireless technologies, rendering it highly applicable to future 6G and high-mobility communication systems.
P-1.6 Holistic Definition of Sustainability Needs and Assessment Methodology
Anastasius Gavras (Eurescom GmbH, Germany); Marja Matinmikko-Blue (University of Oulu, Centre for Wireless Communications, Finland); Julie Bradford (Real Wireless Limited, United Kingdom (Great Britain))
We present ongoing work towards defining and assessing sustainability in the context of 6G mobile networks, moving beyond traditional performance metrics, and introducing environmental, societal, and economic dimensions of sustainability. The work aims to harmonize the concept of sustainability across the mobile network ecosystem, addressing a critical gap in current literature, and facilitate the development of a standardized sustainability assessment methodology for 6G technologies and use cases.
P-1.7 Cellfree Optical Wireless Integrated Sensing and Communication System Design and Test Plan
John Cosmas and Kareem Ali (Brunel University, United Kingdom (Great Britain)); Adam Kapovits and Anastasius Gavras (Eurescom GmbH, Germany); Israel Koffman (RunEL, Israel); Benjamin Azoulay (Oledcomm, France); Clement Lartigue and Emmanuel Plascencia (OLEDCOMM, France); Bastien Béchadergue (University of Versailles Saint-Quentin-en-Yvelines, France & OLEDCOMM, France); Barthélemy Cagneau (Université de Versailles Saint-Quentin-en-Yvelines, France); Luc Chassagne (University of Versailles Saint-Quentin-en-Yvelines, France); Audrey Bienvenu (Eurescom, Germany)
6G Cellfree Optical Wireless Communication network is presented, which is then used to test the performance of Received Signal Strength, Time Difference of Arrival, Angle of Arrival and Fingerprinting localisation.
P-1.8 Flexible Open Architecture and AI-Driven Enabling Technologies for a Novel 6G Connectivity Platform: FLECON-6G
Juan Brenes and Giada Landi (Nextworks, Italy); Christos Verikoukis (University of Patras, Greece); Kostas Ramantas (Iquadrat Informatica, Greece); Adlen Ksentini (Eurecom, France); Loizos Christofi (eBOS Technologies, Cyprus)
The future 6G architecture will evolve into an intelligent, fully adaptive “Network of Networks”. Multiple stakeholders will participate in a new digital ecosystem, contributing different resources and subsystems, even beyond communication technologies. These elements will be dynamically composed and federated through open interfaces to deliver end-to-end services. The FLECON-6G project aims to explore technological enablers to build the foundations for this new 6G Network of Networks model, introducing advanced levels of trustworthiness and automation. Key innovations include intent management, native, distributed and trustworthy AI, zero-touch closed loops with conflict resolution, and Network Digital Twins, all contributing to a pervasive Intelligence Layer. This paper presents the collaborative and distributed framework proposed by FLECON-6G to enable scalable, efficient and sustainable operation of multi-domain and multi-stakeholder programmable networks.
P-1.9 ENSURE-6G: Advancing Security and Privacy with AI in 6G Networks
Charuka Moremada, Vidura Ravihansa and Bartlomiej Siniarski (University College Dublin, 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); Periklis Chatzimisios (International Hellenic University, Greece & University of New Mexico, USA); Gürkan Gür (Zurich University of Applied Sciences (ZHAW), Switzerland); Madhusanka Liyanage (University College Dublin, Ireland)
This paper reviews artificial intelligence (AI) and Machine Learning (ML)-driven security mechanisms in the next-generation 6G networks, analyzing both their advantages and vulnerabilities. It examines emerging AI-based threats, countermeasures, and the sustainability of AI-powered security solutions. Additionally, it explores regulatory and ethical challenges in AI integration. The study highlights key research directions for developing secure and trustworthy AI-driven 6G ecosystems.
P-1.10 MultiX: Advancing 6G-RAN Through Multi-Technology, Multi-Sensor Fusion, Multi-Band and Multi-Static Perception
Pablo Picazo-Martínez and Antonio de la Oliva (Universidad Carlos III de Madrid, Spain)
MultiX integrates sensing and communication in 6G networks through the MultiX Fusion Perceptive 6G-RAN (MP6R) framework, enhancing awareness and adaptability with multi-band, multi-static, and multi-sensor capabilities. Its distributed architecture includes MultiX Perception System (MPS) for sensing, MPRC Controller (MP6RC) for orchestration, and Data Access and Security Hub (DASH) for secure data processing. The Multi-Layer Digital Twin for Industrial Manufacturing showcases its real-time optimization potential, enabling applications like autonomous systems and smart manufacturing.
P-1.11 Machine Learning-Enhanced Physical Layer Key Generation in OFDM Sub-THz Systems
Julio Suarez Gomez (Centro Tecnolóxico de Telecomunicaciones de Galicia – Gradiant,Spain); J. Joaquín Escudero-Garzás (Centro Tecnolóxico de Telecomunicacións de Galicia – Gradiant, Spain); Jorge Pose Eiroa (Centro Tecnolóxico de Telecomunicaciones de Galicia – Gradiant,Spain)
This document presents the first steps and results of a physical layer key generation (PKG) system for sub-THz wireless communications, addressing both Time Division Duplex (TDD) and Frequency Division Duplex (FDD) scenarios. Channel characteristics are extracted using OFDM-based channel sounding, and the FDD operation mode overcomes non-reciprocity challenges by incorporating machine learning techniques to predict downlink characteristics from the uplink measurements. The extracted channel features are processed through quantification methods to minimize the key disagreement rate (KDR).
P-1.12 KQI-Driven Network Slice Resource Configuration
Oswaldo S. Peñaherrera-Pulla, Carlos Baena, Hao Q. Luo-Chen and Jose A. Trujillo (University of Malaga, Spain); Sergio Fortes (University of Málaga, Spain); Raquel Barco (University of Malaga, Spain)
As 5G and Beyond-5G (B5G) networks evolve, managing resources efficiently is crucial for delivering high-quality services. Traditional network slicing methods rely on static Key Performance Indicators (KPIs), often leading to wasted resources and poor user experience. This work introduces a KQI-driven slice configurator that uses machine learning (ML) and optimization techniques to dynamically adjust network slices based on Key Quality Indicators (KQIs). By analyzing real-time network data, adapting to changing conditions, and balancing service quality with resource efficiency, the system ensures optimal performance. Tests in LTE and 5G networks, using 360-degree video streaming, show that this approach improves resource allocation and enhances user experience, demonstrating the potential of intelligent network slicing for future communication networks.
P-1.13 ARMOR: Adversarial Resistance and Model Optimization for Robustness for B5G/6G Open Radio Access Networks
Chamara Sandeepa, Charuka Moremada, Isuru Pinto and Madhusanka Liyanage (University College Dublin, Ireland)
The ARMOR project addresses Artificial Intelligence (AI)-based security challenges in 5G and Beyond (B5G/6G) Open Radio Access Networks (O-RAN) by developing an adversarial testing framework to assess the robustness of AI models. By simulating white-box, black-box, evasion, and poisoning attacks, ARMOR identifies vulnerabilities and evaluates model performance based on accuracy degradation, response time, and resilience. Initial tests are conducted on real 5G O-RAN testbed, extending to external O-RAN platforms for large-scale evaluations. Moreover, Large Language Models (LLMs) are utilized to automate result analysis and provide actionable recommendations, enhancing AI security in B5G/6G networks.
P-1.14 Experimental Validation of RIS-Assisted Dual- Stream Multi-User Communication Testbed for 6G
You-Cheng Chen, Ting-Hao Shih, Kun-Leng Yang, Po-Yi Hsieh and Shih-Cheng Lin (National Chung Cheng University, Taiwan); Chia-Chan Chang (National Chung-Cheng University, Taiwan); Sheng-Fuh Chang (National Chung Cheng University, Taiwan)
A reconfigurable intelligent surface (RIS)-assisted 6G communication testbed is proposed, enabling two data streams with 64-QAM and 16-QAM modulation schemes for two distinct users. The hardware platform integrates an NI Ettus USRP X410 operating at 4.7 GHz with a 1600-cell RIS. First, an OFDMA-based dual-stream waveform is generated, separating downlink payload resources to mitigate interference. Next, the RIS configures its reflecting elements to simultaneously form two beams pointing toward each user, ensuring stable signal reception. The multiple-beam forming approach is adopted for acquiring the demanded phase distribution profile. Finally, experimental tests compare the system’s performance under RIS activated and deactivated scenarios, highlighting significant improvements in error vector magnitude (EVM), bit error rate (BER), and data throughput. Specifically, the dual-user BER increases by approximately 98% with the assistance of RIS.
P-1.15 A System for Monitoring Selected Vital Signs Based on LoRaWAN: a Proof-of-Concept Implementation
Joanna Szewczyk (Poznan University of Technology, Poland & Institute of Computing Science, Poland); Szymon Wilk (Poznan University of Technology & Institute of Computing Science, Poland); Piotr Remlein (Poznan University of Technology & Chair of Wireless Communications, Poland)
We propose a system for remotely monitoring selected human vital signs by leveraging LoRaWAN’s low-power and wide-area capabilities. The system employs multiple wearable sensors to collect selected vital signs, such as heart rate, muscle activity, and skin response, then sends the data using LoRaWAN communications to the two servers — ThingSpeak and The Things Network. The presented proof-of-concept implementation focuses on the possibilities of using long-range communications in a medical wearable sensor system and on the correctness and robustness of the system’s operations.
P-1.16 Radio Frequency Low-Cost Temperature Sensor
Zabdiel Brito-Brito and Jesús Salvador Velázquez-González (Centre Tecnològic de Telecomunicacions de Catalunya (CTTC), Spain); Fermin Mira (Centre Tecnologic de Telecomunicacions de Catalunya (CTTC), Spain); Yi Wang (University of Birmingham, United Kingdom (Great Britain)); Ignacio Llamas-Garro (Centre Tecnològic de Telecomunicacions de Catalunya (CTTC), Spain)
In this work, preliminary results regarding a radio frequency planar low-cost temperature sensor, with potential for sensing and communications integration, are described. The sensor is based on a ring resonator onto which a layer of PDMS is added as the sensor active layer, allowing resonant frequency shift according to varying temperature since the permittivity of the PDMS changes when exposed to different temperatures. The combination of a ring resonator with a layer of PDMS on top allows the development of a thermal sensor using microwave signals, such as those used for communications. The resonant frequency changes at a rate of 1.01 MHz per degree Celsius. The resonant frequency of the ring resonator at room temperature was designed to resonate at 3GHz. This work is proof of concept towards integrated sensing and communication components.
P-1.17 6G-DALI: an End-to-End AI Framework for Automating Data and ML Operations
Marco Ruta and Giacomo Bernini (Nextworks, Italy); Christos Verikoukis (University of Patras, Greece); Vasileios Theodorou (Intracom S.A. Telecom Solutions, Greece); Takai Eddine Kennouche (VIAVI Solutions, France); Adlen Ksentini (Eurecom, France); Luisa Schuhmacher (KU Leuven, Belgium); Kostas Ramantas (Iquadrat Informatica, Greece); Dimitrios Amaxilatis (Spark Works Ltd., Ireland); Carmela Occhipinti (CyberEthics Lab., Italy); Rihan Hai (Technische Universiteit Delft, The Netherlands)
Native support of Artificial Intelligence (AI) and Machine Learning (ML) is one of the pillars of future 6G networks. Despite the opportunities, there are several gaps that are hindering the seamless adoption of AI/ML in 6G. The lack of rich and high-quality datasets needed to train and fine-tune the models, and the need to test and evaluate AI models in a representative staging environment while ensuring ethics and regulatory compliance, represents a challenge without access to an end-to-end 6G testbed or a representative Digital Twin (DT) replica environment. To this end, 6G-DALI, the new SNS-JU Phase-3 project on Reliable AI for 6G Communications Systems and Services, aims at providing an end-to-end AI framework for 6G, structured around two interdependent pillars: AI experimentation as a service via Machine Learning Operations (MLOps) and Data and analytics collection and storage via Data Operations (DataOps). Finally, 6G-DALI will deliver a 6G Dataspace for dataset storage and secure sharing, and a Digital Twin (DT) testbed for data generation on demand.
P-1.18 Extending 3GPP Positioning Protocols for Integrated Sensing and Communication in 6G Networks
Henry Blue (VTT Technical Research Center of Finland, Finland); Stefan Wänstedt (Ericsson Research, Sweden); Tao Chen (VTT Technical Research Centre of Finland LTD, Finland)
We propose a structured procedure and protocol framework to unify sensing and communication within Integrated Sensing and Communication (ISAC) systems. Our approach details a clear, step-by-step sensing process-including capability polling, resource allocation, measurement collection, and data fusion-managed centrally by a Sensing Management Function. Leveraging existing 3GPP positioning protocols (NRPPa and LPP), we identify necessary extensions to fully enable sensing functionalities. By clarifying how current standards can be effectively adapted and pinpointing where additional capabilities are required, we provide a path toward ISAC implementation, streamlining standardization, minimizing complexity, and promoting secure, privacy-conscious sensing for 6G networks.
P-1.19 Radio Wave Diffraction Modeling Using CityGML Vector Building Data
Rimvydas Aleksiejunas and Karolis Stankevičius (Vilnius University, Lithuania)
Estimation of diffraction loss is essential in wireless network planning and optimization, where detailed building data is available. Currently vector-based 3D city models with high level of detail attract a lot of interest. However, the analysis algorithms with vector data are slow and there are no efficient fast numerical methods. The aim of this work is to create fast numerical diffraction algorithm working with high accuracy 3D vector building data. In the proposed model, diffraction loss is estimated according to ITU-R P.526 recommendation using XDraw approximation for fast numerical processing. In this work, performance and accuracy of diffraction prediction model is compared against traditional raster-based XDraw algorithm. The results are calculated using buildings data from 3D BAG open dataset of Amsterdam city.
P-1.20 Dynamic Multilayer Topology Configuration for Energy Optimization
David De la Osa Mostazo (Telefónica, Spain); Pablo Armingol (Telefonica, Spain); Juan Pedro Fernández-Palacios (Telefónica I+D, Spain); Oscar González de Dios (Telefonica I+D, Spain)
This paper presents a dynamic multilayer topology configuration approach for energy optimization in optical networks, orchestrated by a multilayer slice controller managing both IP and optical layers. The proposed solution dynamically transitions between network topologies-such as star during high traffic periods and horseshoe during low traffic periods-based on traffic patterns. Optical cards, transceivers, and equipment are selectively placed into standby mode to reduce energy consumption, ensuring uninterrupted service by proactively reconfiguring traffic (potentially leveraging machine learning techniques) before topology changes. Beyond the day-night model, the approach adapts to real-time or historical traffic data, enabling modular adjustments across multiple network rings. This methodology has the potential to significantly improve energy efficiency while maintaining high network performance, laying the foundation for sustainable next-generation optical networks.
P-1.21 Versatile AI Deployment for 6G Explorations: Simulation Testbed and Illustrative Applications
Mateo Zoughebi, Elliot Cole and Mihai Mitrea (Institut Polytechnique de Paris, France); Hakima Chaouchi (Télécom SudParis, France)
The present study introduces a low-code, template-based testbed for emulating 6G Edge environments. This simulation testbed is designed to cater for the needs of the dual relationship between AI and 6G: AI for 6G (i.e. deploying AI solutions for optimizing 6G network management) and 6G for AI (i.e. leveraging on 6G opportunities for providing enhanced AI services at the edge). First, when it comes to AI for 6G, the testbed proves itself as a versatile tool for deploying and training AI solutions for 6G optimization (with an illustration for congestion prediction) as well as for generating datasets to power further research studies. Secondly, when it comes 6G for AI, the testbed allows various innovative AI usage at the edge to be experiences and benchmarked (with illustration on an aggregated Pytoch-Tensorflow-Pytorch processing task, and with TCP and UDP connections).
P-1.22 Mobilenet Testbed Ecosystem: Meeting 5G, O-RAN and mmWave
Juan José Martín (University of Málaga, Spain); Antonio Tarrias (University of Malaga & Telecommunication Research Institute TELMA, Spain); Sergio Fortes (University of Málaga, Spain); Raquel Barco (University of Malaga, Spain)
The development of cellular network Use Cases (UCs) requires the utilization of real-world equipment to analyze and evaluate its feasibility prior to final implementations. Moreover, there are different technologies, configurations, and architectures that could be candidates to enable each specific UC, which will need to coexist with all the previous. In this context, the Mobilenet Testbed ecosystem provides an effective solution by offering a comprehensive architecture that integrates a wide range of advanced cellular network technologies, including Open- RAN, Fifth Generation (5G) Standalone (SA), and mmWave communications.
P-1.23 SDR and DIFI Based Satellite Channel Emulator
Ionel Petrut (Lasting Software, Romania); Cornel Balint (Lasting Software Timisoara, Romania)
This paper introduces an efficient, versatile, and low-cost SDR based satellite channel emulator for satellite radio channel. The emulator can be also connected using DIFI (Digital Intermediate Frequency Interoperability) to direct access to digital signal and avoid RF modules. A user-friendly graphical interface allows device operation, parameter settings and satellite selection. The emulator allows to select a specific satellite based on Two Line Element (TLE) and derive all the radio channel parameters for emulation. Laboratory and live test with an open solution 5G Non-Terrestrial Network confirm the appropriate emulator operation.
P-1.24 Attributional Life-Cycle Assessment of Wireless Plant Sensing Module
Lyssa Ramaut, Jona Cappelle, Lieven De Strycker and Liesbet Van der Perre (KU Leuven, Belgium)
Understanding how plants interact with their surroundings is crucial, but equally important is evaluating the environmental impact of the electronics used to monitor them. This paper presents a life-cycle assessment (LCA) case study of a plant sensing module designed to track plant growth and enable early detection of plant stress, contributing to improved crop yields, food safety, and quality. The LCA estimates a carbon footprint of 2.7 kgCO2eq for a 10-year sensor lifespan, based on the Global Warming Potential (GWP) over 100 years. A breakdown of the results highlights that the batteries required to sustain operation during this period have the highest environmental impact, followed by integrated circuits (ICs) and the unpopulated printed circuit board (PCB). These findings highlight the importance of sustainable design choices in electronic design. In addition, exploring alternative power sources and improving electronic disposal strategies will significantly reduce the ecological footprint of IoT-based sensing devices.
P-1.25 Federated Learning for Secure and Efficient Device Activity Detection in mMTC Networks
Ali Elkeshawy and Ibrahim Al Ghosh (CentraleSupélec, France); Haifa Farès (Centrale Supélec, France); Amor Nafkha (CentraleSupélec, France)
Grant-free random access in massive machine-type communications enables low-latency connectivity with minimal signaling. However, sporadic device activation requires efficient device activity detection. We propose a federated learning-based device activity detection approach, leveraging distributed training to enhance security and privacy while maintaining low computational complexity. Compared to existing methods, our solution achieves competitive detection performance, addressing scalability and security challenges in mMTC networks.
P-1.26 Design of a 1-Bit Reconfigurable Transmitarray with High Bistatic Radar Cross Section Gain
Ting Hao Shih, You-Cheng Chen, Kun-Leng Yang, Dai-Ting Tsai and Shih-Cheng Lin (National Chung Cheng University, Taiwan); Chia-Chan Chang (National Chung-Cheng University, Taiwan); Sheng-Fuh Chang (National Chung Cheng University, Taiwan)
A 1-bit reconfigurable transmitarray (RTA) with a receive-transmit (RTX) configuration implemented using a 4-layer PCB stackup is proposed by integrating PIN diodes for dynamic phase control. Operating around 11.7 GHz, the RTA achieves a broad 3-dB transmission bandwidth of 23.1%, and a peak measured RCS gain of 56.6 dB at 11.9 GHz. It also supports a wide beam scanning range over 100° (130°-230°). The measured RCS gain pattern closely matches simulation ones, confirming the RTA’s potential for advanced wireless communication applications.
P-1.27 A Multi-Band Full-Duplex Prototype for Integrated Sensing and Communication
Bixing Yan, Andre Kokkeler and Yang Miao (University of Twente, The Netherlands)
Integrated sensing and communication (ISAC) has emerged as a key enabler for the next-generation radio network. ISAC systems aim to enable wireless sensing and data transmission functionality in one integrated system. While existing ISAC prototypes in the literature predominantly focus on single-band full-duplex operations with array antennas and spatially separated sensing targets and communication user equipment (UE) to mitigate signal interference, this paper presents a novel multi-band full-duplex ISAC prototype leveraging a software-defined radio (SDR) USRP X440. In this initial implementation, constrained by host PC performance limitations, the proposed system supports a monostatic sensing link at 24.8 GHz with 150 MHz bandwidth for range estimation alongside a bistatic communication link at 24.2 GHz with 150 MHz bandwidth, each employing distinct orthogonal frequency-division multiplexing (OFDM) waveforms. Experimental results demonstrate successful target detection with a range resolution of 1 m while maintaining robust wireless communication performance with minimal signal interference between sensing and communication functionalities.
P-1.28 AI-Powered Network Digital Twins: Enhancing 6G Intelligence with Active Mechanisms
Zhiheng Yang, Chrysa Papagianni and Adam Belloum (University of Amsterdam, The Netherlands); Sacheendra Talluri (Delft University of Technology & AtLarge Research, The Netherlands); Paola Grosso (University of Amsterdam, The Netherlands); Alexandru Iosup (Vrije Universiteit, The Netherlands)
Network Digital Twins (NDTs) are key to 6G networks, enabling real-time monitoring, analysis, and optimization. This paper introduces the Future Network Service (FNS) AI-based NDTs framework, which integrates predictive analytics, and interactive learning to enable self-correcting and autonomous network management. We also present a case study using uncertainty-driven active learning, demonstrating AI-driven NDTs as fundamental in the realization of self-optimizing 6G networks.
P-1.29 Poster – Towards a Decentralized 5G Core: Seamless Roaming (SimpleRoam) Implementation in Non-3GPP (WiFi) Access Using free5GC
Bilal Ghani and Hakima Chaouchi (Télécom SudParis, France)
The decentralized 5G core network offers high resilience, low latency, scalability & flexibility, and robust connectivity. Radio Access Network (RAN) architecture, such as Open Radio Station (ORS), employs a distributed core network model, where each base station is connected to a different core network. The final objective is to modify/adapt centralized 5G core networks such as free5GC so that when a User Equipment (UE) moves from one ORS to another, it keeps the same IP address for a continued TCP session. We implemented the first steps for this final goal by using the existing Babel routing protocol and Simple Roaming Daemon (sroamd) for seamless roaming of UE between different Access Points in a mesh network connected to free5GC via Non-3GPP Inter-Working Function (N3IWF) – an untrusted Non-3GPP.
P-1.30 Enabling Open Visual Positioning and Discovery for Geospatial Augmented Reality
Gábor Sörös (Nokia Bell Labs (Hungary) & Nokia Solutions and Networks Kft, Hungary); Andor Kovács and Dénes Bisztray (Nokia Bell Labs (Hungary), Hungary); Christine Perey (PEREY Research & Consulting, Switzerland); James Jackson (Open AR Cloud, USA)
We present a key milestone towards open and interoperable geospatial augmented reality. We demonstrate our prototype open visual positioning service, examples of geospatially-indexed content retrieval, and collaboration enabled by sharing user poses and virtual objects in universal geodetic coordinates.
P-1.31 MANTRA-6G: Management and Orchestration for Cloud- and AI-Native 6G Networks
Marios Avgeris (University of Amsterdam, The Netherlands); Nitinder Mohan (Delft University of Technology, The Netherlands); Gergely Pongrácz (Ericsson Research, Hungary); Chrysa Papagianni (University of Amsterdam, The Netherlands); Fernando A. Kuipers (Delft University of Technology, The Netherlands)
Realizing the envisioned capabilities of 6G requires flexibility, programmability, and intelligent automation. To support zero-touch 6G network service management, we propose MANTRA-6G, an integrated cloud-native, AI-native orchestration and management framework that is designed for seamless, robust, scalable, and intelligent service deployment and management across heterogeneous data planes.
P-1.32 Study on Feasibility of Relay-Assisted THz-Communications in an Obstructed LoS Case
Christoph Herold, Carla E. Reinhardt, Annika Sauer and Thomas Kürner (Technische Universität Braunschweig, Germany)
Wireless communications at THz-frequencies can provide high-data rates due to a large amount of available spectrum. Due to challenging propagation characteristics, wireless links need to rely on line-of-sight connections. In this study, the possibility to establish virtual line-of-sight links in a non-line-of-sight scenario is studied using the ray-tracing and link-level-simulation module of the simulation framework SiMoNe. It showed that employing a relay could improve the receive signal power by over 100 dB according to ray-tracing simulations. Link-level simulations based on the predicted multi-path components suggest that an amplify-and-forward or decode-and-forward relay enables communication where it was previously impossible.
P-1.33 Performance Analysis of Sub-Band Full-Duplex Cell-Free Massive MIMO JCAS Systems
Kwadwo Mensah Obeng Afrane, Yang Miao and Andre Kokkeler (University of Twente, The Netherlands)
In-band Full-duplex joint communication and sensing systems require self interference cancellation as well as decoupling of the mutual interference between UL communication signals and radar echoes. We present sub-band full-duplex as an alternative duplexing scheme to achieve simultaneous uplink communication and target parameter estimation in a cell-free massive MIMO system. Sub-band full-duplex allows uplink and downlink transmissions simultaneously on non-overlapping frequency resources via explicitly defined uplink and downlink sub-bands in each timeslot. Thus, we propose a sub-band full-duplex cell-free massive MIMO system with active downlink sensing on downlink sub-bands and uplink communication on uplink sub-band. In the proposed system, the target illumination signal is transmitted on the downlink (radar) sub-band whereas uplink users transmit on the uplink (communication) sub-band. By assuming efficient suppression of inter-sub-band interference between radar and communication sub-bands, uplink communication and radar signals can be efficiently processed without mutual interference. We show that each AP can estimate sensing parameters with high accuracy in SBFD cell-free massive MIMO JCAS systems.
P-1.34 Evaluating the Energy Consumption of Secure HTTP Video Streaming over 5G
Jere Koivisto (VTT, Finland); Mikko Uitto and Jukka Mäkelä (VTT Technical Research Centre of Finland Ltd, Finland)
The increasing global consumption of online video content necessitates research into reducing the energy consumption of video streaming. This study evaluates the energy efficiency of the HTTP protocol under various conditions, focusing on the performance improvements in different HTTP and TLS protocol versions and the impact of video buffer sizes on energy consumption. Empirical measurements were conducted across multiple components of the video streaming architecture. The results indicate that HTTP/3.0, which uses QUIC as its transport protocol, is the most energy-efficient version for video streaming. Additionally, longer video buffers can significantly reduce the overall power consumption of user equipment (UE). These findings suggest that adopting HTTP/3.0 with optimal video buffer lengths can enhance the sustainability of video streaming services.
P-1.35 Deep Learning for Efficient Antenna Spacing Control in Beamforming Antennas
Kenta Umebayashi (Tokyo University of Agriculture and Technology, Japan); Antti Tölli (University of Oulu, Finland)
This study proposes a Deep Learning (DL)-based antenna spacing control method for Adaptive Antenna Arrays with Weight and Spacing control (AAA-WS) to efficiently suppress interference signals. Conventional Adaptive Antenna Arrays based on Weight control (AAA-W) adjust only the weights while keeping antenna spacing fixed, requiring M+1 antennas to suppress M interference signals. In contrast, AAA-WS achieves effective interference suppression using only two antennas by optimizing antenna spacing in a line-of-sight environment. However, previous AAA-WS methods relied on exhaustive search, requiring direction-of-arrival (DOA) information and imposing high computational costs. To address these issues, we propose a DL-based approach that determines optimal spacing using received signal correlation matrices instead of DOA. By refining training data constraints, we develop an efficient and lightweight DL model for adaptive antenna spacing control.
P-1.36 On the Assistance of Unmanned Aerial Vehicle to Wireless Cellular Systems
Vicente Casares-Giner (Universitat Politècnica de València, Spain); Xiaohu Ge and Yuxi Zhao (Huazhong University of Science and Technology, China)
In this paper we consider the collaboration of Unmanned Aerial Vehicles (UAV) with wireless cellular mobile systems. UAVs cooperate closely with base stations (BSs) when they are occasionally present in the coverage area of one BS. From the tele-traffic point of view, UAVs can provide additional capacity to cellular BSs such as to alleviate saturation conditions during high congested periods. The assignment of traffic channels works as follows; when a call arrives to the system it is assigned to any free channel of the BS. If all channels of the BS are busy, the call is assigned to any free channel of any present UAV. If all channels of the present UAVs are busy, the call is lost. When a call served by a given BS ends, any other call in progress on a UAV, if any, is transferred to the released channel of that BS. This strategy prevents possible forced termination due to abandonment or departure of the UAV from the coverage area of the referred BS. The scenario under study is modeled as a 2-D Markov process in which we evaluate the blocking probability of new calls, the probability of forced termination of ongoing calls, and due to the end of an ongoing call at the BS, the probability of handover or reassignment of an ongoing call at a UAV to the BS.
