Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 38
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

AI-powered edge computing evolution for beyond 5G communication networks

  • Kartsakli, Elli
  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Bartzoudis, Nikolaos
  • Frascolla, Valerio
  • Mohalik, Swarup Kumar
  • Metsch, Thijs
  • Antonopoulos, Angelos|||0000-0002-3546-1080
  • Tuna, Ömer Faruk
  • Deng, Yansha
  • Tao, Xin
  • Serrano, Maria A.
  • Quiñones Moreno, Eduardo
Edge computing is a key enabling technology that is expected to play a crucial role in beyond 5G (B5G) and 6G communication networks. By bringing computation closer to where the data is generated, and leveraging Artificial Intelligence (AI) capabilities for advanced automation and orchestration, edge computing can enable a wide range of emerging applications with extreme requirements in terms of latency and computation, across multiple vertical domains. In this context, this paper first discusses the key technological challenges for the seamless integration of edge computing within B5G/6G and then presents a roadmap for the edge computing evolution, proposing a novel design approach for an open, intelligent, trustworthy, and distributed edge architecture., VERGE has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Advanced edge computing architecture for AI-driven automation and slicing in beyond 5G

  • Kartsakli, Elli
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Pujol Roig, Joan
  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Bartzoudis, Nikolaos
  • Gülen, Utku
  • Kosu, Semiha
  • Kolawole, Oluwatayo
  • Tao, Xin
  • Mohalik, Swarup Kumar
  • Antonopoulos, Angelos|||0000-0002-3546-1080
  • Kalalas, Charalampos
  • Quiñones Moreno, Eduardo
This paper proposes an open, modular, and distributed architecture for the edge-to-cloud compute continuum, in synergy with the beyond 5G and 6G architectural trends. The proposed design facilitates the lifecycle management of cloud-native services, as well as Artificial Intelligence (AI) pipelines developed for the optimization of complex and heterogeneous edge-enabled environments, while supporting a suite of novel mechanisms to ensure the trustworthiness of AI solutions. Finally, the proposed architecture introduces a potential extension of the network slicing paradigm for the joint provisioning of communication and computing resources for applications running at the edge., Project VERGE has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. UK participants in VERGE are supported by UKRI grant numbers 10071211 (Samsung Electronics (UK) Limited) and 10061781 (King’s College London)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

On exploiting user equipment relaying capabilities in beyond 5G networks: opportunities, challenges, and road map

  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Ruiz García, Olga
Mobile Network Operators are facing significant expenditures in deploying 5G Radio Access Network (RAN) infrastructure to achieve the level of densification required for meeting the new services demands. This will be exacerbated in Beyond 5G scenarios, which envisage the operation in high frequency bands with poor propagation. Therefore, operators need to find new and cost-efficient ways of managing and deploying their infrastructure. In this context, sustained on the technological evolution of the User Equipment (UE) and on recent standardization initiatives, this paper proposes an upgraded Beyond 5G system based on augmenting the RAN through the exploitation of UE-to-network relaying capabilities for optimizing the coverage footprint and for bringing edge computing capabilities closer to the users. The paper presents the baseline enabling technologies for realizing this concept. discusses the key architectural requirements and challenges for each one and presents a possible implementation roadmap, supported by some evaluation results, which reflect that relaying can improve the throughput in a factor 6 and reduce outage probability in around 80%., This article is part of the ARTIST project (Reference PID2020-115104RB-I00), funded by the Spanish Ministry of Science and Innovation, under Grant MCIN/AEI/10.13039/501100011033. The article is also part of the VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking, under the European Union’s Horizon Europe research and innovation program, under Grant 101096034., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Towards trustworthy reinforcement learning-based resource management in beyond 5G

  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Kartsakli, Elli
  • Tuna, Ömer Faruk
  • Mohalik, Swarup Kumar
  • Tao, Xin
It is envisaged that future Beyond 5G (B5G) systems will make extensive use of Artificial Intelligence (AI) capabilities to achieve an efficient automated management and optimization of communication and computing resources and to support advanced data-driven applications that provide the users with highly immersive services. Ensuring the trustworthiness of the AI solutions is key for their successful introduction in B5G, as this will guarantee their robustness towards errors and potential attack threats, the privacy of the used data and trained models and the explainability of AI-based decisions, ensuring that they do not have unsafe consequences. In this context, this paper focuses on the trustworthiness of Reinforcement Learning (RL) solutions, as their inherent trial-and-error behavior during training makes them particularly challenging from the robustness perspective. Then, the paper proposes a framework for managing the lifecycle of an RL-based resource management solution for both training and inference stages to ensure its trustworthy operation. The framework relies on an RL training configuration function to specify the training conditions, a Network Digital Twin (NDT) to perform the training on a safe environment and a continuous operation function to monitor the behaviour of the trained policy during inference. The framework is illustrated with an applicability use case of capacity sharing for network slicing., This work is part of VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. The work of Irene Vilà has been funded by a Margarita Salas Grant (ref. 2022UPC-MSC- 94079)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Attention-driven AI model generalization for workload forecasting in the compute continuum

  • Gort Jelmer Dirk, Berend
  • Kibalya, Godfrey Mirondo|||0000-0002-7053-3756
  • Antonopoulos, Angelos|||0000-0002-3546-1080
Effective resource management in edge-cloud networks demands precise forecasting of diverse workload resource usage. Due to the fluctuating nature of user demands, prediction models must have strong generalization abilities, ensuring high performance amidst sudden traffic changes or unfamiliar patterns. Existing approaches often struggle with handling long-term dependencies and the diversity of temporal patterns. This paper introduces OmniFORE (Framework for Optimization of Resource forecasts in Edge-cloud networks), which integrates attention-based time-series models with temporal clustering to enhance generalization and predict diverse workloads efficiently in volatile settings. By training on carefully selected subsets from extensive datasets, OmniFORE captures both short-term stability and long-term shifts in resource usage patterns. Experiments show that OmniFORE outperforms state-of-the-art methods in prediction accuracy, inference speed, and generalization to unseen data, particularly in scenarios with dynamic workload changes and varying trace variance. These improvements enable more efficient resource management in the compute continuum., This work was supported in part by Horizon Europe through the research project VERGE under Grant SNS-JU-101096034; and in part by the Universalización de Infraestructuras Digitales para la Cohesión - Plan de Recuperación, Transformación y Resiliencia (UNICO-5G-RPTR) through the project AEON-ZERO under Grant TSI-063000-2021-52, through the project FREE-6G under Grant TSI-063000-2021-144, through the project AROMA3D under Grant TSI-063000-2021-70/71, through the project 6G-OASIS under Grant TSI-063000-2021-24, through the project SUCCESS-6G under Grant TSI-063000-2021-39/40/41, and through the project AVANZANDO-5G-GEMELOS DIGITALES under Grant TSI-063000-2021-112/113/114., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

On the deployment of an AI-driven power control mechanism for D-MIMO in beyond 5G scenarios

  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Tuna, Ömer Faruk
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Bartzoudis, Nikolaos
  • Kartsakli, Elli
  • Kolawole, Oluwatayo
  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Serrano, Maria A.
  • Gülen, Utku
The evolution beyond the Fifth Generation (5G) of mobile communications systems is expected to exploit different advanced radio access techniques, such as distributed multiple input multiple output (D-MIMO), together with an extensive support of Artificial Intelligence (AI) and Machine Learning (ML) solutions for different aspects of network optimization. In addition, these systems will also benefit from the evolution of edge computing technologies through the availability of computing and storage capabilities that will span from the radio access network nodes to the cloud, forming an edge-to-cloud compute continuum. The availability of these features will provide the beyond 5G and 6G networks with an increased flexibility for deploying the different AI/ML-based optimization functionalities, trading-off aspects such as the delay requirements, the computational complexity or the availability of data for feeding the AI/ML models. In this context, this paper takes as a reference the evolved edge computing architecture of the VERGE project and focuses on the possibilities to deploy on top it an AI/ML model for D-MIMO power control. The paper discusses several options to host the associated AI/ML model lifecycle management functionalities on the edge-to-cloud compute continuum and elaborates on the implications of the considered approach in terms of standardization., This work is part of VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. Samsung Research UK participants in Horizon Europe Project VERGE are supported by UKRI grant number 10071211. The work of Ericsson Research Turkey authors is also partly funded by The Scientific and Technological Research Council of Turkey (TUBITAK) through the 1515 Frontier Research and Development Laboratories Support Program under Project 5169902., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

A transfer reinforcement learning approach for capacity sharing in Beyond 5G networks

  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
The use of Reinforcement Learning (RL) techniques has been widely addressed in the literature to cope with capacity sharing in 5G Radio Access Network (RAN) slicing. These algorithms consider a training process to learn an optimal capacity sharing decision-making policy, which is later applied to the RAN environment during the inference stage. When relevant changes occur in the RAN, such as the deployment of new cells in the network, RL-based capacity sharing solutions require a re-training process to update the optimal decision-making policy, which may require long training times. To accelerate this process, this paper proposes a novel Transfer Learning (TL) approach for RL-based capacity sharing solutions in multi-cell scenarios that is implementable following the Open-RAN (O-RAN) architecture and exploits the availability of computing resources at the edge for conducting the training/inference processes. The proposed approach allows transferring the weights of the previously learned policy to learn the new policy to be used after the addition of new cells. The performance assessment of the TL solution highlights its capability to reduce the training process duration of the policies when adding new cells. Considering that the roll-out of 5G networks will continue for several years, TL can contribute to enhancing the practicality and feasibility of applying RL-based solutions for capacity sharing., This work has been partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No. 101096034 (VERGE project). Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. This research has also been partly funded by the Spanish Ministry of Economic Affairs and Digital Transformation and the European Union—NextGenerationEU under projects OPTIMAIX_OaaS (Ref. TSI-063000- 2021-34) and OPTIMAIX_NDT (Ref. TSI-063000-2021-35). The work of Irene Vilà has been funded by the European Union-NextGenerationEU, the Spanish Ministry of Universities, and the Plan for Recovery, Transformation, and Resilience, through the call for Margarita Salas Grants of the Universitat Politècnica de Catalunya (ref. 2022UPC-MSC-94079)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

An evolutionary edge computing architecture for beyond 5G era

  • Kartsakli, Elli
  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Bartzoudis, Nikolaos
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Kolawole, Oluwatayo
  • Tao, Xin
  • Mohalik, Swarup Kumar
  • Mach, Tomasz
  • Liu, Sige
  • Deng, Yansha
  • Mandò, Gianluca
  • Antonopoulos, Angelos|||0000-0002-3546-1080
  • Frascolla, Valerio
  • Kosu, Semiha
  • Kalem, Gokhan
  • Buining, Fred
  • Quiñones Moreno, Eduardo
Beyond 5G (B5G) communication networks face the challenge of meeting the demanding requirements of various service types, including uRLLC, mIoT, eMBB, and emerging technologies like Extended Reality (XR). Edge computing can address these demands effectively because of the ability to bring computational power and resources closer to the source of data. Nevertheless, the realization of this potential necessitates an open, flexible, and automated architectural framework capable of supporting disaggregated applications and network designs. In this context, this paper introduces a novel architecture designed to advance the evolution of edge computing in B5G, developed within the EU-funded project VERGE. The proposed architecture is modular and scalable, guided by artificial intelligence (AI), and founded on three essential pillars: "edge for AI," "AI for edge," and "security, privacy, and trustworthiness for AI." After presenting this architecture, the paper showcases its applicability through the examination of two vertical use cases within the industrial and transportation domains., Project VERGE has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. UK participants in VERGE are supported by UKRI grant numbers 10071211 (Samsung Electronics (UK) Limited) and 10061781 (King’s College London), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Expanding edge computing deeper into beyond 5G radio access networks

  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Pérez Romero, Jordi|||0000-0001-9131-5013
Relevant services envisaged for beyond 5G (B5G) systems, such as extended reality and holographic communications, present stringent user experience requirements with high computational and communication demands. While edge computing aims to address the computation requirements by offloading the computational tasks to edge servers close to the user, the communication will leverage the technologies developed for 5G New Radio together with an unprecedented level of network densification. This paper advocates for deploying relays equipped with edge computing capabilities. The potentials of this approach for B5G are identified and a system model is presented to characterize both computational and communications perspectives. Based on this, results are provided to show the benefits and limitations of the proposed approach from a system-level perspective., This work is part of VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking under the European Union's Horizon Europe research and innovation programme under Grant Agreement 101096034. Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. This paper has also been partly funded by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020-115104RB-I00). The work of Irene Vilà has been funded by European UnionNextGenerationEU, Spanish Ministry of Universities and the Plan for Recovery, Transformation and Resilience, through the call for Margarita Salas Grants of the Universitat Politècnica de Catalunya (ref. 2022UPC-MSC- 94079)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

ReLVaaS: verification-as-a-service to analyze trustworthiness of RL-based solutions in 6G networks

  • Tao, Xin
  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Mohalik, Swarup Kumar
  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
Reinforcement learning (RL) based solutions are emerging as key enablers for integrating Artificial Intelligence (AI) in future 6G networks due to their ability to tackle complex decision-making and control problems. However, current testing-based validation and evaluation methods for these models may not guarantee their trustworthiness, thereby hindering their adoption in real-world applications, especially for safety or mission-critical scenarios. In contrast, formal verification is presented as a promising solution, but its effective use in RL verification requires key steps of property specification, effective system model construction, and framework support for different analyses. Towards this, we present ReLVaaS, Reinforcement Learning Verification-as-a-Service. ReLVaaS provides templates to facilitate the specification of complex trustworthiness properties, requirement-guided system model construction, and invoking off-the-shelf model checking engines, all in an automated process. ReLVaaS is demonstrated on a DRL-based (Deep Reinforcement Learning) capacity sharing solution for Radio Access Network (RAN) slicing, showcasing the capabilities of probabilistic model checking in providing quantitative analysis of trustworthiness properties of resilience, safety, and reliability., This work is part of the VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. The work of Irene Vila has been funded by a Margarita Salas Grant (ref. 2022UPC-MSC- 94079). The expressed views and opinions are, however, those of the authors only and may not reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Space and time user distribution measurements dataset in a university campus

  • Ruiz García, Olga
  • Sánchez González, Juan|||0000-0002-2415-9628
  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Vilà Muñoz, Irene|||0000-0002-7086-9591
Radio Resource Management (RRM) strategies are essential components in mobile wireless networks, such as Long Term Evolution (LTE) or 5G New Radio (NR)[1]. The design of RRM algorithmic solutions is an area of research that has received a lot of attention for decades (see e.g. surveys [2–4] and references therein). Given that RRM strategies need to handle traffic dynamicity in the Radio Access Network (RAN), the availability of realistic user distributions in space and time can be very useful to support a more realistic performance assessment of RRM solutions. For this purpose, this article introduces a dataset containing real measurements of the number of users connected to the different Wifi Access Points (APs) at the Campus Nord facilities of the Universitat Politècnica de Catalunya (UPC) in Barcelona. The Wifi network is composed of 247 APs. To characterize the temporal variations, the data was collected for each AP every 1000 s (approximately) during 62 days. Besides the number of users connected to each AP (including users connected to both 2.4 GHz and 5 GHz bands), the dataset also contains information about the theoretical coverage area of each AP, so that the number of users connected to each AP can be associated to a specific geographical area. In this way, the dataset captures the spatio-temporal variations of users in the Campus at different times of the day, different days of the week and different periods of the academic year., This work is partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreements No. 101096034 (VERGE project) and No. 101097083 (BeGREEN project) and by the Spanish Ministry of Science and Innovation MCIN/AEI/ 10.13039/501100011033 under ARTIST project (ref. PID2020–115104RB-I00). Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. The work of Irene Vila has been funded by the European Union-NextGenerationEU, the Spanish Ministry of Universities and the Plan for Recovery, Transformation and Resilience, through the call for Margarita Salas Grants of the Universitat Politecnica de Catalunya (ref. 2022UPC-MSC- 94079)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Low layer functional split management in 5G and beyond: architecture and self-adaptation

  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Campoy García, David|||0009-0004-3135-3834
  • Gelonch Bosch, Antonio José|||0000-0001-9216-1140
  • Gelabert Doran, Xavier
  • Klaiqi, Bleron
Radio Access Network (RAN) disaggregation is emerging as a key trend in beyond 5G, as it offers new opportunities for more flexible deployments and intelligent network management. A relevant problem in disaggregated RAN is the functional split selection, which dynamically decides which baseband (BB) functions of a base station are kept close to the radio units and which ones are centralized. In this context, this paper firstly presents an architectural framework for supporting this concept relying on the O-RAN architecture. Then, the paper analyzes how the functional split can be optimized to adapt to the different load conditions while minimizing energy costs., The work by J. Pérez-Romero, O. Sallent and D. Campoy has been partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No. 101096034 (VERGE project), by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020-115104RB-I00) and by Huawei Technologies. The work by X. Gelabert and B. Klaiqi has been partially funded by the European Union's Horizon Europe research and innovation programme (HORIZON-MSCA-2021-DN-0) under the Marie Skłodowska-Curie grant agreement No. 101073265., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Digital twin as a service for 6G radio access networks: Functional model and key challenges

  • Campoy García, David|||0009-0004-3135-3834
  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
This paper presents a scalable and flexible functional model for the 6G-Radio Access Network Digital Twin (6G-RANDT), designed through a service-oriented approach to support next-generation radio access network optimization and intelligent management. The proposed model includes an application programming interface for seamless interaction with a 6G-RANDT consumer, 6G-RANDT instances for RAN emulation and simulation, and a 6G-RANDT orchestrator for management and data integration. Several 6G-RANDT typologies and measurement requirements are explored to address diverse use cases. Moreover, key challenges in 6GRANDT development are discussed. To evaluate the impact of modeling accuracy in 6G-RANDTs, we conducted a cellular capacity modeling experiment in which different 6G-RANDT instances were used to train a reinforcement learning (RL)-based capacity-sharing algorithm. Results show that the impact of capacity modeling strategies on the 6G-RANDT instance accuracy and RL agent performance is highly dependent on the cell environment, emphasizing the need for tailored 6G-RANDT services based on specific RAN environment conditions and applications., This work has been partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No. 101096034 (VERGE project), grant ARTIST (PID2020-115104RB-I00) and TRAINER-6G (PID2023-146748OB-I00) funded by MCIN/AEI/10.13039/501100011033 and by ERDF/EU., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Knowledge transfer for collaborative misbehavior detection in untrusted vehicular environments

  • Sedar, Mohottige Roshan Madhusanka|||0000-0003-3170-5575
  • Kalalas, Charalampos
  • Dini, Paolo
  • Vazquez-Gallego, Francisco
  • Alonso Zárate, Jesús
  • Alonso Zárate, Luis Gonzaga|||0000-0002-6608-0862
© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Vehicular mobility underscores the need for collaborative misbehavior detection at the vehicular edge. However, locally trained misbehavior detection models are susceptible to adversarial attacks that aim to deliberately influence learning outcomes. In this paper, we introduce a deep reinforcement learning-based approach that employs transfer learning for collaborative misbehavior detection among roadside units (RSUs). In the presence of label-flipping and policy induction attacks, we perform selective knowledge transfer from trustworthy source RSUs to foster relevant expertise in misbehavior detection and avoid negative knowledge sharing from adversary-influenced RSUs. The performance of our proposed scheme is demonstrated with evaluations over a diverse set of misbehavior detection scenarios using an open-source dataset. Experimental results show that our approach significantly reduces the training time at the target RSU and achieves superior detection performance compared to the baseline scheme with tabula rasa learning. Enhanced robustness and generalizability can also be attained, by effectively detecting previously unseen and partially observable misbehavior attacks., This work has been partially funded by the ”Ministerio de Asuntos Economicos y Transformación Digital” and the European Union- NextGenerationEU in the frameworks of the ”Plan de Recuperacion, Transformacion y Resiliencia” and the ”Mecanismo de Recuperación y Resiliencia under references TSI-063000-2021-39, TSI-063000-2021-40, TSI-063000- 2021-41, and by the European Commission’s Horizon Europe, Smart Networks and Services Joint Undertaking research and innovation program, under project VERGE (GA No. 101096034)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

A new methodology for user equipment trajectory prediction in cellular networks

  • Sánchez González, Juan|||0000-0002-2415-9628
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Pérez Romero, Jordi|||0000-0001-9131-5013
User mobility prediction can be exploited in cellular networks for different purposes, such as enhancing the handover process, proactive resource allocation, proactive load balancing, etc., in order to improve the network performance. While many works aimed for the prediction of the next cell visited by the User Equipment (UE), the prediction of future UE locations has received less attention. In fact, only a few works deal with the prediction of the next UE location while other few works aim to predict the future direction of UEs arriving at a crossroad. This paper presents a methodology for the prediction of the trajectory followed by the UEs inside the cell. First, UE trajectory patterns are learnt by means of an off-line clustering of historical UE trajectories. Then, the obtained trajectory patterns are used for on-line prediction of the UE trajectory inside the cell, the prediction of the next cell that the UE will visit and an estimation of the time to reach this new cell. A dataset with UE trajectories moving around a large real-life cellular network has been considered., This work is part of VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking under the European Union’s Horizon Europe research and innovation programme under Grant Agreement 101096034. This paper has also been partly funded by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020- 115104RB-I00, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

On the deployment of AI-driven solutions in edge-enabled beyond 5G scenarios

  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Tuna, Ömer Faruk
  • Kolawole, Oluwatayo
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Bartzoudis, Nikolaos
  • Kartsakli, Elli
  • Pujol Roig, Joan
  • Serrano, Maria A.
  • Gülen, Utku
The evolution beyond the Fifth Generation (5G) of mobile communications systems is expected to exploit different advanced radio access techniques, such as distributed multiple input multiple output (D-MIMO) or network slicing, together with an extensive support of Artificial Intelligence (AI) and Machine Learning (ML) solutions for different aspects of network optimization. These systems will also benefit from the evolution of edge computing technologies through the availability of computing and storage capabilities that will span from the radio access network nodes to the cloud, forming an edge-to-cloud compute continuum. The availability of these features will provide the beyond 5G (B5G) and 6G networks with an increased flexibility for deploying the different AI/ML-based optimization functionalities, trading-off aspects such as delay requirements, computational complexity or availability of data for feeding the AI/ ML models. In this context, this paper takes as a reference the evolved edge computing architecture of the VERGE project and presents a pipeline with the required functionalities for the lifecycle management of AI/ML models in edge-enabled B5G systems. This pipeline is particularized for the deployment of two specific AI-driven solutions used for capacity sharing in network slicing and for D-MIMO power control., This work was supported in part by the VERGE Project from the Smart Networks and Services Joint Undertaking (SNS JU) through the European Union’s Horizon Europe Research and Innovation Programme under Grant 101096034; and in part by the Samsung Research U.K. participants in Horizon Europe Project VERGE through UKRI under Grant 10071211. The work of Ömer Faruk Tuna and Utku Gülen was supported in part by the Scientific and Technological Research Council of Turkey (TUBITAK) through the 1515 Frontier Research and Development Laboratories Support Program under Project 5169902. The work of Jordi Pérez-Romero, Oriol Sallent, and Irene Vilà was supported by the TRAINER-6G Project through MCIN/ AEI/10.13039/501100011033 and ERDF/EU under Grant PID2023-146748OBI00. The work of Nikolaos Bartzoudis was supported in part by MICIU/ AEI/10.13039/501100011033 and the European Union NextGenerationEU/PRTR under Project PID2023- 146245OB-C22. The work of Elli Kartsakli was supported in part by the PROXIMITY Project through AEI/10.13039/501100011 033/FEDER, UE, under Grant PID2021-124122OA-I00., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

A tutorial on the characterisation and modelling of low layer functional splits for flexible radio access networks in 5G and beyond

  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Gelonch Bosch, Antonio José|||0000-0001-9216-1140
  • Gelabert Doran, Xavier
  • Klaiqi, Bleron
  • Kahn, Marcus
  • Campoy García, David|||0009-0004-3135-3834
The centralization of baseband (BB) functions in a radio access network (RAN) towards data processing centres is receiving increasing interest as it enables the exploitation of resource pooling and statistical multiplexing gains among multiple cells, facilitates the introduction of collaborative techniques for different functions (e.g., interference coordination), and more efficiently handles the complex requirements of advanced features of the fifth generation (5G) new radio (NR) physical layer, such as the use of massive multiple input multiple output (MIMO). However, deciding the functional split (i.e., which BB functions are kept close to the radio units and which BB functions are centralized) embraces a trade-off between the centralization benefits and the fronthaul costs for carrying data between distributed antennas and data processing centres. Substantial research efforts have been made in standardization fora, research projects and studies to resolve this trade-off, which becomes more complicated when the choice of functional splits is dynamically achieved depending on the current conditions in the RAN. This paper presents a comprehensive tutorial on the characterisation, modelling and assessment of functional splits in a flexible RAN to establish a solid basis for the future development of algorithmic solutions of dynamic functional split optimisation in 5G and beyond systems. First, the paper explores the functional split approaches considered by different industrial fora, analysing their equivalences and differences in terminology. Second, the paper presents a harmonized analysis of the different BB functions at the physical layer and associated algorithmic solutions presented in the literature, assessing both the computational complexity and the associated performance. Based on this analysis, the paper presents a model for assessing the computational requirements and fronthaul bandwidth requirements of different functional splits. Last, the model is used to derive illustrative results that identify the major trade-offs that arise when selecting a functional split and the key elements that impact the requirements., This work has been partially funded by Huawei Technologies. Work by X. Gelabert and B. Klaiqi is partially funded by the European Union's Horizon Europe research and innovation programme (HORIZON-MSCA-2021-DN-0) under the Marie Skłodowska-Curie grant agreement No 101073265. Work by J. Perez-Romero and O. Sallent is also partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreements No. 101096034 (VERGE project) and No. 101097083 (BeGREEN project) and by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020-115104RB-I00). This last project has also funded the work by D. Campoy., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Deep Learning-based algorithm for optimizing relay user equipment activation in 5G cellular networks

  • Hernández Carlón, Juan Jesús
  • Pérez Romero, Jordi|||0000-0001-9131-5013
  • Sallent Roig, Oriol|||0000-0002-2114-1406
  • Vilà Muñoz, Irene|||0000-0002-7086-9591
  • Casadevall Palacio, Fernando José|||0000-0001-5385-872X
This paper addresses the problem of optimally using the relay capabilities of user equipment (UE) to augment the radio access network (RAN) in 5G deployments and beyond. This can be particularly useful in coverage constrained scenarios, such as those using millimeter waves, due to the difficulty radio signals penetrate some structures. This can lead to signal blockages and high penetration losses when providing outdoor-to-indoor coverage. To overcome these limitations, the use of relay UEs (RUEs) is seen as a possible solution to effectively extend the coverage of a cellular network. In this context, this paper proposes a deep learning-based algorithm to optimize the decision regarding when RUEs should be activated and deactivated in accordance with the benefits they can provide for increasing the spectral efficiency and decreasing outage probability for the network users. The obtained results reveal a promising capability of the proposed solution to activate the most beneficial RUEs given the network conditions being experienced, leading to improvements of average spectral efficiency of 12.3% and reductions of outage probability of 89% with respect to the case without relays., FUNDING: This work was supported in part by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreements No. 101096034 (VERGE project) and No. 101097083 (BeGREEN project) and in part by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020-115104RB-I00). The work of J.J. Hernández-Carlón was supported by the Spanish Ministry of Science and Innovation under grant ref. PRE2018-084691. The work of I. Vilà was supported by the European Union-NextGenerationEU, Spanish Ministry of Universities and the Plan for Recovery,, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

OmniFORE: Attention-based generalization framework for edge-cloud workload predictions

  • Gort Jelmer Dirk, Berend
  • Kibalya, Godfrey Mirondo|||0000-0002-7053-3756
  • Umbert Juliana, Anna|||0000-0001-7825-5212
  • Antonopoulos, Angelos|||0000-0002-3546-1080
Effective resource management in edge-cloud networks requires accurate prediction of resource utilization across diverse workloads. The dynamic nature of user demands necessitates prediction models with strong generalization capabilities, i.e., able to achieve high performance in sudden traffic changes or even unseen patterns. Existing works struggle with long-term dependencies and varied temporal patterns. This paper proposes OmniFORE (Framework for Optimization of Resource forecasts in Edge-cloud networks). In particular, we combine attention-based time-series models with temporal clustering to achieve robust generalization and efficiently consume and predict diverse workloads in volatile environments. By training on representative subsets from extensive datasets, OmniFORE captures both short-term stability and long-term changes in resource usage patterns. Experiments with real-world data show that our approach outperforms state-of-the-art methods, particularly in new environments, improving prediction accuracy by 83.87% and achieving 17.92% faster inference., This work is supported by the VERGE project (SNS-JU101096034) under Horizon Europe and the AVANZANDO5G-GEMELOS DIGITALES project (TSI-063000-2021-112/113/114) under UNICO-5G-RPTR., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A new methodology for user equipment trajectory prediction in cellular networks

  • Sánchez González, Juan
  • Sallent Roig, Oriol
  • Pérez Romero, Jordi
User mobility prediction can be exploited in cellular networks for different purposes, such as enhancing the handover process, proactive resource allocation, proactive load balancing, etc., in order to improve the network performance. While many works aimed for the prediction of the next cell visited by the User Equipment (UE), the prediction of future UE locations has received less attention. In fact, only a few works deal with the prediction of the next UE location while other few works aim to predict the future direction of UEs arriving at a crossroad. This paper presents a methodology for the prediction of the trajectory followed by the UEs inside the cell. First, UE trajectory patterns are learnt by means of an off-line clustering of historical UE trajectories. Then, the obtained trajectory patterns are used for on-line prediction of the UE trajectory inside the cell, the prediction of the next cell that the UE will visit and an estimation of the time to reach this new cell. A dataset with UE trajectories moving around a large real-life cellular network has been considered., This work is part of VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking under the European Union’s Horizon Europe research and innovation programme under Grant Agreement 101096034. This paper has also been partly funded by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020- 115104RB-I00, Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

On the deployment of an AI-driven power control mechanism for D-MIMO in beyond 5G scenarios

  • Pérez Romero, Jordi
  • Tuna, Ömer Faruk
  • Sallent Roig, Oriol
  • Bartzoudis, Nikolaos
  • Kartsakli, Elli
  • Kolawole, Oluwatayo
  • Vilà Muñoz, Irene
  • Serrano, Maria A.
  • Gülen, Utku
The evolution beyond the Fifth Generation (5G) of mobile communications systems is expected to exploit different advanced radio access techniques, such as distributed multiple input multiple output (D-MIMO), together with an extensive support of Artificial Intelligence (AI) and Machine Learning (ML) solutions for different aspects of network optimization. In addition, these systems will also benefit from the evolution of edge computing technologies through the availability of computing and storage capabilities that will span from the radio access network nodes to the cloud, forming an edge-to-cloud compute continuum. The availability of these features will provide the beyond 5G and 6G networks with an increased flexibility for deploying the different AI/ML-based optimization functionalities, trading-off aspects such as the delay requirements, the computational complexity or the availability of data for feeding the AI/ML models. In this context, this paper takes as a reference the evolved edge computing architecture of the VERGE project and focuses on the possibilities to deploy on top it an AI/ML model for D-MIMO power control. The paper discusses several options to host the associated AI/ML model lifecycle management functionalities on the edge-to-cloud compute continuum and elaborates on the implications of the considered approach in terms of standardization., This work is part of VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. Samsung Research UK participants in Horizon Europe Project VERGE are supported by UKRI grant number 10071211. The work of Ericsson Research Turkey authors is also partly funded by The Scientific and Technological Research Council of Turkey (TUBITAK) through the 1515 Frontier Research and Development Laboratories Support Program under Project 5169902., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Expanding edge computing deeper into beyond 5G radio access networks

  • Vilà Muñoz, Irene
  • Sallent Roig, Oriol
  • Pérez Romero, Jordi
Relevant services envisaged for beyond 5G (B5G) systems, such as extended reality and holographic communications, present stringent user experience requirements with high computational and communication demands. While edge computing aims to address the computation requirements by offloading the computational tasks to edge servers close to the user, the communication will leverage the technologies developed for 5G New Radio together with an unprecedented level of network densification. This paper advocates for deploying relays equipped with edge computing capabilities. The potentials of this approach for B5G are identified and a system model is presented to characterize both computational and communications perspectives. Based on this, results are provided to show the benefits and limitations of the proposed approach from a system-level perspective., This work is part of VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking under the European Union's Horizon Europe research and innovation programme under Grant Agreement 101096034. Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. This paper has also been partly funded by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020-115104RB-I00). The work of Irene Vilà has been funded by European UnionNextGenerationEU, Spanish Ministry of Universities and the Plan for Recovery, Transformation and Resilience, through the call for Margarita Salas Grants of the Universitat Politècnica de Catalunya (ref. 2022UPC-MSC- 94079)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Low layer functional split management in 5G and beyond: architecture and self-adaptation

  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
  • Campoy García, David
  • Gelonch Bosch, Antonio José
  • Gelabert Doran, Xavier
  • Klaiqi, Bleron
Radio Access Network (RAN) disaggregation is emerging as a key trend in beyond 5G, as it offers new opportunities for more flexible deployments and intelligent network management. A relevant problem in disaggregated RAN is the functional split selection, which dynamically decides which baseband (BB) functions of a base station are kept close to the radio units and which ones are centralized. In this context, this paper firstly presents an architectural framework for supporting this concept relying on the O-RAN architecture. Then, the paper analyzes how the functional split can be optimized to adapt to the different load conditions while minimizing energy costs., The work by J. Pérez-Romero, O. Sallent and D. Campoy has been partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No. 101096034 (VERGE project), by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020-115104RB-I00) and by Huawei Technologies. The work by X. Gelabert and B. Klaiqi has been partially funded by the European Union's Horizon Europe research and innovation programme (HORIZON-MSCA-2021-DN-0) under the Marie Skłodowska-Curie grant agreement No. 101073265., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Deep Learning-based algorithm for optimizing relay user equipment activation in 5G cellular networks

  • Hernández Carlón, Juan Jesús
  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
  • Vilà Muñoz, Irene
  • Casadevall Palacio, Fernando José
This paper addresses the problem of optimally using the relay capabilities of user equipment (UE) to augment the radio access network (RAN) in 5G deployments and beyond. This can be particularly useful in coverage constrained scenarios, such as those using millimeter waves, due to the difficulty radio signals penetrate some structures. This can lead to signal blockages and high penetration losses when providing outdoor-to-indoor coverage. To overcome these limitations, the use of relay UEs (RUEs) is seen as a possible solution to effectively extend the coverage of a cellular network. In this context, this paper proposes a deep learning-based algorithm to optimize the decision regarding when RUEs should be activated and deactivated in accordance with the benefits they can provide for increasing the spectral efficiency and decreasing outage probability for the network users. The obtained results reveal a promising capability of the proposed solution to activate the most beneficial RUEs given the network conditions being experienced, leading to improvements of average spectral efficiency of 12.3% and reductions of outage probability of 89% with respect to the case without relays., FUNDING: This work was supported in part by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreements No. 101096034 (VERGE project) and No. 101097083 (BeGREEN project) and in part by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020-115104RB-I00). The work of J.J. Hernández-Carlón was supported by the Spanish Ministry of Science and Innovation under grant ref. PRE2018-084691. The work of I. Vilà was supported by the European Union-NextGenerationEU, Spanish Ministry of Universities and the Plan for Recovery,, Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Space and time user distribution measurements dataset in a university campus

  • Ruiz García, Olga
  • Sánchez González, Juan
  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
  • Vilà Muñoz, Irene
Radio Resource Management (RRM) strategies are essential components in mobile wireless networks, such as Long Term Evolution (LTE) or 5G New Radio (NR)[1]. The design of RRM algorithmic solutions is an area of research that has received a lot of attention for decades (see e.g. surveys [2–4] and references therein). Given that RRM strategies need to handle traffic dynamicity in the Radio Access Network (RAN), the availability of realistic user distributions in space and time can be very useful to support a more realistic performance assessment of RRM solutions. For this purpose, this article introduces a dataset containing real measurements of the number of users connected to the different Wifi Access Points (APs) at the Campus Nord facilities of the Universitat Politècnica de Catalunya (UPC) in Barcelona. The Wifi network is composed of 247 APs. To characterize the temporal variations, the data was collected for each AP every 1000 s (approximately) during 62 days. Besides the number of users connected to each AP (including users connected to both 2.4 GHz and 5 GHz bands), the dataset also contains information about the theoretical coverage area of each AP, so that the number of users connected to each AP can be associated to a specific geographical area. In this way, the dataset captures the spatio-temporal variations of users in the Campus at different times of the day, different days of the week and different periods of the academic year., This work is partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreements No. 101096034 (VERGE project) and No. 101097083 (BeGREEN project) and by the Spanish Ministry of Science and Innovation MCIN/AEI/ 10.13039/501100011033 under ARTIST project (ref. PID2020–115104RB-I00). Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. The work of Irene Vila has been funded by the European Union-NextGenerationEU, the Spanish Ministry of Universities and the Plan for Recovery, Transformation and Resilience, through the call for Margarita Salas Grants of the Universitat Politecnica de Catalunya (ref. 2022UPC-MSC- 94079)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Attention-driven AI model generalization for workload forecasting in the compute continuum

  • Gort Jelmer Dirk, Berend
  • Kibalya, Godfrey Mirondo
  • Antonopoulos, Angelos
Effective resource management in edge-cloud networks demands precise forecasting of diverse workload resource usage. Due to the fluctuating nature of user demands, prediction models must have strong generalization abilities, ensuring high performance amidst sudden traffic changes or unfamiliar patterns. Existing approaches often struggle with handling long-term dependencies and the diversity of temporal patterns. This paper introduces OmniFORE (Framework for Optimization of Resource forecasts in Edge-cloud networks), which integrates attention-based time-series models with temporal clustering to enhance generalization and predict diverse workloads efficiently in volatile settings. By training on carefully selected subsets from extensive datasets, OmniFORE captures both short-term stability and long-term shifts in resource usage patterns. Experiments show that OmniFORE outperforms state-of-the-art methods in prediction accuracy, inference speed, and generalization to unseen data, particularly in scenarios with dynamic workload changes and varying trace variance. These improvements enable more efficient resource management in the compute continuum., This work was supported in part by Horizon Europe through the research project VERGE under Grant SNS-JU-101096034; and in part by the Universalización de Infraestructuras Digitales para la Cohesión - Plan de Recuperación, Transformación y Resiliencia (UNICO-5G-RPTR) through the project AEON-ZERO under Grant TSI-063000-2021-52, through the project FREE-6G under Grant TSI-063000-2021-144, through the project AROMA3D under Grant TSI-063000-2021-70/71, through the project 6G-OASIS under Grant TSI-063000-2021-24, through the project SUCCESS-6G under Grant TSI-063000-2021-39/40/41, and through the project AVANZANDO-5G-GEMELOS DIGITALES under Grant TSI-063000-2021-112/113/114., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Towards trustworthy reinforcement learning-based resource management in beyond 5G

  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
  • Vilà Muñoz, Irene
  • Kartsakli, Elli
  • Tuna, Ömer Faruk
  • Mohalik, Swarup Kumar
  • Tao, Xin
It is envisaged that future Beyond 5G (B5G) systems will make extensive use of Artificial Intelligence (AI) capabilities to achieve an efficient automated management and optimization of communication and computing resources and to support advanced data-driven applications that provide the users with highly immersive services. Ensuring the trustworthiness of the AI solutions is key for their successful introduction in B5G, as this will guarantee their robustness towards errors and potential attack threats, the privacy of the used data and trained models and the explainability of AI-based decisions, ensuring that they do not have unsafe consequences. In this context, this paper focuses on the trustworthiness of Reinforcement Learning (RL) solutions, as their inherent trial-and-error behavior during training makes them particularly challenging from the robustness perspective. Then, the paper proposes a framework for managing the lifecycle of an RL-based resource management solution for both training and inference stages to ensure its trustworthy operation. The framework relies on an RL training configuration function to specify the training conditions, a Network Digital Twin (NDT) to perform the training on a safe environment and a continuous operation function to monitor the behaviour of the trained policy during inference. The framework is illustrated with an applicability use case of capacity sharing for network slicing., This work is part of VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. The work of Irene Vilà has been funded by a Margarita Salas Grant (ref. 2022UPC-MSC- 94079)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A transfer reinforcement learning approach for capacity sharing in Beyond 5G networks

  • Vilà Muñoz, Irene
  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
The use of Reinforcement Learning (RL) techniques has been widely addressed in the literature to cope with capacity sharing in 5G Radio Access Network (RAN) slicing. These algorithms consider a training process to learn an optimal capacity sharing decision-making policy, which is later applied to the RAN environment during the inference stage. When relevant changes occur in the RAN, such as the deployment of new cells in the network, RL-based capacity sharing solutions require a re-training process to update the optimal decision-making policy, which may require long training times. To accelerate this process, this paper proposes a novel Transfer Learning (TL) approach for RL-based capacity sharing solutions in multi-cell scenarios that is implementable following the Open-RAN (O-RAN) architecture and exploits the availability of computing resources at the edge for conducting the training/inference processes. The proposed approach allows transferring the weights of the previously learned policy to learn the new policy to be used after the addition of new cells. The performance assessment of the TL solution highlights its capability to reduce the training process duration of the policies when adding new cells. Considering that the roll-out of 5G networks will continue for several years, TL can contribute to enhancing the practicality and feasibility of applying RL-based solutions for capacity sharing., This work has been partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No. 101096034 (VERGE project). Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. This research has also been partly funded by the Spanish Ministry of Economic Affairs and Digital Transformation and the European Union—NextGenerationEU under projects OPTIMAIX_OaaS (Ref. TSI-063000- 2021-34) and OPTIMAIX_NDT (Ref. TSI-063000-2021-35). The work of Irene Vilà has been funded by the European Union-NextGenerationEU, the Spanish Ministry of Universities, and the Plan for Recovery, Transformation, and Resilience, through the call for Margarita Salas Grants of the Universitat Politècnica de Catalunya (ref. 2022UPC-MSC-94079)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

On the deployment of AI-driven solutions in edge-enabled beyond 5G scenarios

  • Pérez Romero, Jordi
  • Tuna, Ömer Faruk
  • Kolawole, Oluwatayo
  • Sallent Roig, Oriol
  • Vilà Muñoz, Irene
  • Bartzoudis, Nikolaos
  • Kartsakli, Elli
  • Pujol Roig, Joan
  • Serrano, Maria A.
  • Gülen, Utku
The evolution beyond the Fifth Generation (5G) of mobile communications systems is expected to exploit different advanced radio access techniques, such as distributed multiple input multiple output (D-MIMO) or network slicing, together with an extensive support of Artificial Intelligence (AI) and Machine Learning (ML) solutions for different aspects of network optimization. These systems will also benefit from the evolution of edge computing technologies through the availability of computing and storage capabilities that will span from the radio access network nodes to the cloud, forming an edge-to-cloud compute continuum. The availability of these features will provide the beyond 5G (B5G) and 6G networks with an increased flexibility for deploying the different AI/ML-based optimization functionalities, trading-off aspects such as delay requirements, computational complexity or availability of data for feeding the AI/ ML models. In this context, this paper takes as a reference the evolved edge computing architecture of the VERGE project and presents a pipeline with the required functionalities for the lifecycle management of AI/ML models in edge-enabled B5G systems. This pipeline is particularized for the deployment of two specific AI-driven solutions used for capacity sharing in network slicing and for D-MIMO power control., This work was supported in part by the VERGE Project from the Smart Networks and Services Joint Undertaking (SNS JU) through the European Union’s Horizon Europe Research and Innovation Programme under Grant 101096034; and in part by the Samsung Research U.K. participants in Horizon Europe Project VERGE through UKRI under Grant 10071211. The work of Ömer Faruk Tuna and Utku Gülen was supported in part by the Scientific and Technological Research Council of Turkey (TUBITAK) through the 1515 Frontier Research and Development Laboratories Support Program under Project 5169902. The work of Jordi Pérez-Romero, Oriol Sallent, and Irene Vilà was supported by the TRAINER-6G Project through MCIN/ AEI/10.13039/501100011033 and ERDF/EU under Grant PID2023-146748OBI00. The work of Nikolaos Bartzoudis was supported in part by MICIU/ AEI/10.13039/501100011033 and the European Union NextGenerationEU/PRTR under Project PID2023- 146245OB-C22. The work of Elli Kartsakli was supported in part by the PROXIMITY Project through AEI/10.13039/501100011 033/FEDER, UE, under Grant PID2021-124122OA-I00., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A tutorial on the characterisation and modelling of low layer functional splits for flexible radio access networks in 5G and beyond

  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
  • Gelonch Bosch, Antonio José
  • Gelabert Doran, Xavier
  • Klaiqi, Bleron
  • Kahn, Marcus
  • Campoy García, David
The centralization of baseband (BB) functions in a radio access network (RAN) towards data processing centres is receiving increasing interest as it enables the exploitation of resource pooling and statistical multiplexing gains among multiple cells, facilitates the introduction of collaborative techniques for different functions (e.g., interference coordination), and more efficiently handles the complex requirements of advanced features of the fifth generation (5G) new radio (NR) physical layer, such as the use of massive multiple input multiple output (MIMO). However, deciding the functional split (i.e., which BB functions are kept close to the radio units and which BB functions are centralized) embraces a trade-off between the centralization benefits and the fronthaul costs for carrying data between distributed antennas and data processing centres. Substantial research efforts have been made in standardization fora, research projects and studies to resolve this trade-off, which becomes more complicated when the choice of functional splits is dynamically achieved depending on the current conditions in the RAN. This paper presents a comprehensive tutorial on the characterisation, modelling and assessment of functional splits in a flexible RAN to establish a solid basis for the future development of algorithmic solutions of dynamic functional split optimisation in 5G and beyond systems. First, the paper explores the functional split approaches considered by different industrial fora, analysing their equivalences and differences in terminology. Second, the paper presents a harmonized analysis of the different BB functions at the physical layer and associated algorithmic solutions presented in the literature, assessing both the computational complexity and the associated performance. Based on this analysis, the paper presents a model for assessing the computational requirements and fronthaul bandwidth requirements of different functional splits. Last, the model is used to derive illustrative results that identify the major trade-offs that arise when selecting a functional split and the key elements that impact the requirements., This work has been partially funded by Huawei Technologies. Work by X. Gelabert and B. Klaiqi is partially funded by the European Union's Horizon Europe research and innovation programme (HORIZON-MSCA-2021-DN-0) under the Marie Skłodowska-Curie grant agreement No 101073265. Work by J. Perez-Romero and O. Sallent is also partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreements No. 101096034 (VERGE project) and No. 101097083 (BeGREEN project) and by the Spanish Ministry of Science and Innovation MCIN/AEI/10.13039/501100011033 under ARTIST project (ref. PID2020-115104RB-I00). This last project has also funded the work by D. Campoy., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

AI-powered edge computing evolution for beyond 5G communication networks

  • Kartsakli, Elli
  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
  • Bartzoudis, Nikolaos
  • Frascolla, Valerio
  • Mohalik, Swarup Kumar
  • Metsch, Thijs
  • Antonopoulos, Angelos
  • Tuna, Ömer Faruk
  • Deng, Yansha
  • Tao, Xin
  • Serrano, Maria A.
  • Quiñones Moreno, Eduardo
Edge computing is a key enabling technology that is expected to play a crucial role in beyond 5G (B5G) and 6G communication networks. By bringing computation closer to where the data is generated, and leveraging Artificial Intelligence (AI) capabilities for advanced automation and orchestration, edge computing can enable a wide range of emerging applications with extreme requirements in terms of latency and computation, across multiple vertical domains. In this context, this paper first discusses the key technological challenges for the seamless integration of edge computing within B5G/6G and then presents a roadmap for the edge computing evolution, proposing a novel design approach for an open, intelligent, trustworthy, and distributed edge architecture., VERGE has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

On exploiting user equipment relaying capabilities in beyond 5G networks: opportunities, challenges, and road map

  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
  • Ruiz García, Olga
Mobile Network Operators are facing significant expenditures in deploying 5G Radio Access Network (RAN) infrastructure to achieve the level of densification required for meeting the new services demands. This will be exacerbated in Beyond 5G scenarios, which envisage the operation in high frequency bands with poor propagation. Therefore, operators need to find new and cost-efficient ways of managing and deploying their infrastructure. In this context, sustained on the technological evolution of the User Equipment (UE) and on recent standardization initiatives, this paper proposes an upgraded Beyond 5G system based on augmenting the RAN through the exploitation of UE-to-network relaying capabilities for optimizing the coverage footprint and for bringing edge computing capabilities closer to the users. The paper presents the baseline enabling technologies for realizing this concept. discusses the key architectural requirements and challenges for each one and presents a possible implementation roadmap, supported by some evaluation results, which reflect that relaying can improve the throughput in a factor 6 and reduce outage probability in around 80%., This article is part of the ARTIST project (Reference PID2020-115104RB-I00), funded by the Spanish Ministry of Science and Innovation, under Grant MCIN/AEI/10.13039/501100011033. The article is also part of the VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking, under the European Union’s Horizon Europe research and innovation program, under Grant 101096034., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Advanced edge computing architecture for AI-driven automation and slicing in beyond 5G

  • Kartsakli, Elli
  • Sallent Roig, Oriol
  • Pujol Roig, Joan
  • Pérez Romero, Jordi
  • Vilà Muñoz, Irene
  • Bartzoudis, Nikolaos
  • Gülen, Utku
  • Kosu, Semiha
  • Kolawole, Oluwatayo
  • Tao, Xin
  • Mohalik, Swarup Kumar
  • Antonopoulos, Angelos
  • Kalalas, Charalampos
  • Quiñones Moreno, Eduardo
This paper proposes an open, modular, and distributed architecture for the edge-to-cloud compute continuum, in synergy with the beyond 5G and 6G architectural trends. The proposed design facilitates the lifecycle management of cloud-native services, as well as Artificial Intelligence (AI) pipelines developed for the optimization of complex and heterogeneous edge-enabled environments, while supporting a suite of novel mechanisms to ensure the trustworthiness of AI solutions. Finally, the proposed architecture introduces a potential extension of the network slicing paradigm for the joint provisioning of communication and computing resources for applications running at the edge., Project VERGE has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. UK participants in VERGE are supported by UKRI grant numbers 10071211 (Samsung Electronics (UK) Limited) and 10061781 (King’s College London)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Digital twin as a service for 6G radio access networks: Functional model and key challenges

  • Campoy García, David
  • Vilà Muñoz, Irene
  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
This paper presents a scalable and flexible functional model for the 6G-Radio Access Network Digital Twin (6G-RANDT), designed through a service-oriented approach to support next-generation radio access network optimization and intelligent management. The proposed model includes an application programming interface for seamless interaction with a 6G-RANDT consumer, 6G-RANDT instances for RAN emulation and simulation, and a 6G-RANDT orchestrator for management and data integration. Several 6G-RANDT typologies and measurement requirements are explored to address diverse use cases. Moreover, key challenges in 6GRANDT development are discussed. To evaluate the impact of modeling accuracy in 6G-RANDTs, we conducted a cellular capacity modeling experiment in which different 6G-RANDT instances were used to train a reinforcement learning (RL)-based capacity-sharing algorithm. Results show that the impact of capacity modeling strategies on the 6G-RANDT instance accuracy and RL agent performance is highly dependent on the cell environment, emphasizing the need for tailored 6G-RANDT services based on specific RAN environment conditions and applications., This work has been partially funded by the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No. 101096034 (VERGE project), grant ARTIST (PID2020-115104RB-I00) and TRAINER-6G (PID2023-146748OB-I00) funded by MCIN/AEI/10.13039/501100011033 and by ERDF/EU., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Knowledge transfer for collaborative misbehavior detection in untrusted vehicular environments

  • Sedar, Mohottige Roshan Madhusanka
  • Kalalas, Charalampos
  • Dini, Paolo
  • Vazquez-Gallego, Francisco
  • Alonso Zárate, Jesús
  • Alonso Zárate, Luis Gonzaga
© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Vehicular mobility underscores the need for collaborative misbehavior detection at the vehicular edge. However, locally trained misbehavior detection models are susceptible to adversarial attacks that aim to deliberately influence learning outcomes. In this paper, we introduce a deep reinforcement learning-based approach that employs transfer learning for collaborative misbehavior detection among roadside units (RSUs). In the presence of label-flipping and policy induction attacks, we perform selective knowledge transfer from trustworthy source RSUs to foster relevant expertise in misbehavior detection and avoid negative knowledge sharing from adversary-influenced RSUs. The performance of our proposed scheme is demonstrated with evaluations over a diverse set of misbehavior detection scenarios using an open-source dataset. Experimental results show that our approach significantly reduces the training time at the target RSU and achieves superior detection performance compared to the baseline scheme with tabula rasa learning. Enhanced robustness and generalizability can also be attained, by effectively detecting previously unseen and partially observable misbehavior attacks., This work has been partially funded by the ”Ministerio de Asuntos Economicos y Transformación Digital” and the European Union- NextGenerationEU in the frameworks of the ”Plan de Recuperacion, Transformacion y Resiliencia” and the ”Mecanismo de Recuperación y Resiliencia under references TSI-063000-2021-39, TSI-063000-2021-40, TSI-063000- 2021-41, and by the European Commission’s Horizon Europe, Smart Networks and Services Joint Undertaking research and innovation program, under project VERGE (GA No. 101096034)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

OmniFORE: Attention-based generalization framework for edge-cloud workload predictions

  • Gort Jelmer Dirk, Berend
  • Kibalya, Godfrey Mirondo
  • Umbert Juliana, Anna
  • Antonopoulos, Angelos
Effective resource management in edge-cloud networks requires accurate prediction of resource utilization across diverse workloads. The dynamic nature of user demands necessitates prediction models with strong generalization capabilities, i.e., able to achieve high performance in sudden traffic changes or even unseen patterns. Existing works struggle with long-term dependencies and varied temporal patterns. This paper proposes OmniFORE (Framework for Optimization of Resource forecasts in Edge-cloud networks). In particular, we combine attention-based time-series models with temporal clustering to achieve robust generalization and efficiently consume and predict diverse workloads in volatile environments. By training on representative subsets from extensive datasets, OmniFORE captures both short-term stability and long-term changes in resource usage patterns. Experiments with real-world data show that our approach outperforms state-of-the-art methods, particularly in new environments, improving prediction accuracy by 83.87% and achieving 17.92% faster inference., This work is supported by the VERGE project (SNS-JU101096034) under Horizon Europe and the AVANZANDO5G-GEMELOS DIGITALES project (TSI-063000-2021-112/113/114) under UNICO-5G-RPTR., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

An evolutionary edge computing architecture for beyond 5G era

  • Kartsakli, Elli
  • Pérez Romero, Jordi
  • Bartzoudis, Nikolaos
  • Sallent Roig, Oriol
  • Kolawole, Oluwatayo
  • Tao, Xin
  • Mohalik, Swarup Kumar
  • Mach, Tomasz
  • Liu, Sige
  • Deng, Yansha
  • Mandò, Gianluca
  • Antonopoulos, Angelos
  • Frascolla, Valerio
  • Kosu, Semiha
  • Kalem, Gokhan
  • Buining, Fred
  • Quiñones Moreno, Eduardo
Beyond 5G (B5G) communication networks face the challenge of meeting the demanding requirements of various service types, including uRLLC, mIoT, eMBB, and emerging technologies like Extended Reality (XR). Edge computing can address these demands effectively because of the ability to bring computational power and resources closer to the source of data. Nevertheless, the realization of this potential necessitates an open, flexible, and automated architectural framework capable of supporting disaggregated applications and network designs. In this context, this paper introduces a novel architecture designed to advance the evolution of edge computing in B5G, developed within the EU-funded project VERGE. The proposed architecture is modular and scalable, guided by artificial intelligence (AI), and founded on three essential pillars: "edge for AI," "AI for edge," and "security, privacy, and trustworthiness for AI." After presenting this architecture, the paper showcases its applicability through the examination of two vertical use cases within the industrial and transportation domains., Project VERGE has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. UK participants in VERGE are supported by UKRI grant numbers 10071211 (Samsung Electronics (UK) Limited) and 10061781 (King’s College London), Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

ReLVaaS: verification-as-a-service to analyze trustworthiness of RL-based solutions in 6G networks

  • Tao, Xin
  • Vilà Muñoz, Irene
  • Mohalik, Swarup Kumar
  • Pérez Romero, Jordi
  • Sallent Roig, Oriol
Reinforcement learning (RL) based solutions are emerging as key enablers for integrating Artificial Intelligence (AI) in future 6G networks due to their ability to tackle complex decision-making and control problems. However, current testing-based validation and evaluation methods for these models may not guarantee their trustworthiness, thereby hindering their adoption in real-world applications, especially for safety or mission-critical scenarios. In contrast, formal verification is presented as a promising solution, but its effective use in RL verification requires key steps of property specification, effective system model construction, and framework support for different analyses. Towards this, we present ReLVaaS, Reinforcement Learning Verification-as-a-Service. ReLVaaS provides templates to facilitate the specification of complex trustworthiness properties, requirement-guided system model construction, and invoking off-the-shelf model checking engines, all in an automated process. ReLVaaS is demonstrated on a DRL-based (Deep Reinforcement Learning) capacity sharing solution for Radio Access Network (RAN) slicing, showcasing the capabilities of probabilistic model checking in providing quantitative analysis of trustworthiness properties of resilience, safety, and reliability., This work is part of the VERGE project, which has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under the European Union’s Horizon Europe research and innovation programme under Grant Agreement No 101096034. The work of Irene Vila has been funded by a Margarita Salas Grant (ref. 2022UPC-MSC- 94079). The expressed views and opinions are, however, those of the authors only and may not reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them., Peer Reviewed, Postprint (published version)