Search

The evolution towards 6G will require changes in the way how networks have to be operated to face the stringent service demands while at the same time using efficiently the available resources. In this context, this paper presents the Optimisation-as-a-Service (OaaS) platform being designed by the OPTIMAIX project that makes use of the Network Digital Twin (NDT) concept to test network configurations in a controlled environment. After describing the OaaS platform architecture, the paper presents the optimisation algorithms considered by the project and summarizes the use cases developed in the prototyping activities., This work is 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. TSI063000-2021-35)., Peer Reviewed, Postprint (published version)

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)

The continuous evolution of the communications networks in the race to support 6G services will introduce important operational challenges for network operators, especially in terms of efficient usage of resources, near real-time performance, etc. Such advanced 6G services will necessarily leverage on existing brownfield networks complemented with novel technologies and supportive tools. In this context, the concept of network should be understood in a broad sense, including both networking and computing capabilities enabling the connectivity of the customers to service functions or applications in a very dynamic way. Until recently, the main driver for networks design and operation has been the guarantee of throughput since bandwidth-based services have been dominant so far. With the advent of 5G, delay-based services have started to become relevant as well, as a consequence of the offerings of low latency and interactive services, such as tactile internet or augmented and virtual reality. Now, the advent of forthcoming beyond 5G (B5G) and 6G services is expected to introduce more stringent requirements motivating a new radical change on operations, adding new dimensions to service provision, such as delay-variation or precision-based services. All this complexity triggers the need of evolving the mode of operation by introducing smarter, faster, and educated decisions on the operational processes, pursuing an overall optimization on the usage of both networking and compute resources. In this context, the concept of Network Digital Twins (NDTs) is gaining traction as a perfect complement to the foreseen complex operation., This work is 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. TSI063000-2021-35)., Peer Reviewed, Postprint (author's final draft)

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)

Relevant services envisaged for beyond 5G (B5G) systems, such as extended reality and holographic communications, have extremely demanding user experience requirements with significant computational and communication demands. While edge computing aims to address the computation requirements by offloading the computational tasks to edge servers near the user, the communication will take advantage of the technologies developed for 5G New Radio jointly with an never-before-seen degree of network densification. This paper proposes the use of relays with edge computing capabilities. The approach's potential for B5G are identified, and a system model is defined to characterize both computational and communications viewpoints. Based on this, results are provided to highlight the gains and limitations of the proposed approach from a system-level perspective. Finally, the main challenges for enabling relays with computing capabilities in B5G deployments are discussed., 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 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 Politècnica de Catalunya (ref. 2022UPC-MSC- 94079)., Peer Reviewed, Postprint (published version)

High-speed trains (HST) are nowadays more present in our lives currently, some of them can reach speeds up to 500 km/h and futuristic concepts such as hyperloop tunnels could make trains travel at speeds up to 1000 km/h. Dealing with such high speeds arises many communication problems, for example, in mobility management, with many handovers or high Doppler frequency shifts. You might be thinking how it is possible to provide a good QoS to the users inside the train, when traveling at such elevated velocities. In the thesis, we rely on the development of 5G New Radio and the benefits associated, such as a new handover protocol introduced by 3GPP called conditional handover (CHO). By simulating with Simu5G a HST scenario we have proved that CHO can provide a better service to the users by improving the SINR levels and being more efficient than common handover., Los trenes de alta velocidad están cada vez más presentes en nuestro día a día, algunos ya alcanzan velocidades de 500 km/h, mientras que otros conceptos futuristas como los túneles hyperloop podrían hacer que alcanzaran velocidades de hasta 1000 km/h. En el ámbito de las telecomunicaciones, trabajar a tan altas velocidades conlleva algunos problemas, como por ejemplo un elevado número de handovers. Seguramente, os estéis preguntando cómo es posible establecer un servicio que cumpla unos mínimos de calidad para el usuario, cuando este viaja a tan altas velocidades. Para ello, nos hemos apoyado en la tecnología 5G i un nuevo concepto de handover llamado conditional handover (CHO), introducido por el 3GPP. A través del simulador Simu5G, hemos conseguido demostrar que el CHO no solo es un protocolo más eficiente, sino que además conlleva una mejora en los niveles de SINR, en condiciones parecidas a las de un tren de alta velocidad., Els trens d'alta velocitat estan cada vegada més presents en el nostre dia a dia, alguns ja son capaços d'arribar a velocitats pròximes als 500 km/h, mentre que altres conceptes futuristes com els túnels hyperloop podrien fer que els trens arribessin a velocitats de 1000 km/h. En l'àmbit de les comunicacions, treballar amb velocitats tan elevades comporta alguns problemes, com per exemple un ampli número de handovers. Segurament, estareu pensant com es possible establir un servei que compleixi uns mínims de qualitat de cara a l'usuari, al estar treballant amb velocitats tant elevades. Per fer-ho ens hem recolzat en la tecnologia 5G i un nou concepte de handover presentat pel 3GPP, el conditional handover (CHO). Simulant a través de Simu5G un escenari similar al d'un tren d'alta velocitat, hem pogut demostrar que el CHO no es només un protocol més eficient que el handover normal, sinó que a més a més millora els nivells de SINR.

A Network Digital Twin (NDT) is a high-fidelity digital mirror of a real network. Given the increasing complexity of 5G and beyond networks, the use of an NDT becomes useful as a platform for testing configurations and algorithms prior to their application in the real network, as well as for predicting the performance of such algorithms under different conditions. While an NDT can be defined for the different subsystems of the network, this paper proposes an NDT architecture focusing on the Radio Access Network (RAN), describing the components to represent and model the operation of the different RAN elements, and to perform emulations. Different application use cases are identified, and among them, the paper puts the focus on the training of Reinforcement Learning (RL) solutions for the RAN. For this use case, the paper introduces a framework aligned with O-RAN specifications and discusses the functionalities needed to integrate the NDT. This use case is illustrated with the description of a RAN NDT implementation used for training an RL-based capacity-sharing solution for network slicing. Presented results demonstrate that the implemented RAN NDT is a suitable platform to successfully train the RL solution, achieving service-level agreement satisfaction values above 85%., Peer Reviewed, Postprint (published version)

Capacity sharing in RAN slicing, The 5G network slicing feature allows the creation of a multiple of logical networks, known as "network slices", on top of a shared physical infrastructure. Each network slice is customized to specific service requirements and operates in an separate manner from other slices. This technology also is difficult to apply in Radio Access Networks (RANs), since it demands efficient radio resource management while meeting specific needs for each created RAN slices. The first part of this thesis provides an overview of the DRL technology and its application to RAN Management challenges. This will help to develop advanced solutions for capacity sharing in 5G networks. Then, a DRL-based solution is evaluated for multi-tenant scenarios within 5G RAN. The solution involves a Multi-Agent Reinforcement Learning (MARL) approach with Deep Q-Network (DQN) agents. DQN's MARL-based solution is designed to address capacity-sharing challenges within multi-cell networks, allowing users to adjust to changing traffic conditions while maintaining Service Level Agreements (SLA) for each network segment. To this end, the solution is evaluated in four different scenarios and its strengths, weaknesses and opportunities for improvement are examined. Collab platform is used as a tool to test the potential of this tool as an education tool.

The advent of 5G has meant providing users with higher data rates, more connected devices and lower latency, implying the need to redefine traditional network management methods. In addition, 6G is expected to provide much more challenging and restrictive performance, enhancing the idea of redefining network management methodologies, aimed at maximising the use of network resources. Network Digital Twinning (NDT) and Optimization-as-a-Service (OaaS) are two key mechanisms, along with the use of artificial intelligence (AI) and machine learning (ML), to assist in network management procedures. On the one hand, the NDT is a digital replica of the physical network, which permits evaluating the impact of concrete configuration settings prior to its real implementation in the system, helping to optimize the overall network performance. On the other hand, OaaS offers ad-hoc optimized recommendations to third-party applications based on the execution of several optimization algorithms. In this work, we propose an open, flexible and scalable OaaS platform, defined in the OPTIMAIX project, to exploit the NDT potential targeted to support automated planning and optimization of networks. Also, an evaluation of the NDT role within this architecture is provided., This work is 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. TSI063000-2021-35)., Peer Reviewed, Postprint (published version)

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)

This paper proposes a model to detect and solve coverage holes in 5G Radio Access Network (RAN) deployments operating with millimeter waves. The proposed model utilizes a DBSCAN-based detector to identify coverage constrained areas and then proposes the use of Relay User Equipment (RUE) capabilities to extend the RAN coverage. To optimize the activation and deactivation of RUEs, a Deep-Q-Network-based algorithm is proposed, aiming to improve spectral efficiency and decrease the outage probability experienced by network users. The obtained results demonstrate the effectiveness of the model in accurately detecting coverage constrained areas and efficiently solving these issues by means of an effective RUE activation, leading to significant improvements in network performance while minimizing the time that RUEs remain in active mode, which implies potential benefits for MNOs and UE holders and significant energy savings., This paper is part of ARTIST project (ref. PID2020- 115104RB-I00) funded by MCIN/AEI/10.13039/ 501100011033. The work is also funded by the Spanish Ministry of Science and Innovation under grant ref. PRE2018-084691. The work of Irene Vilà has also been funded by European Union-NextGenerationEU, 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)

One of the applications where fifth generation (5G) networks are expected to have a greatest impact is vehicular-to-everything (V2X) communications. The exchange of data among different vehicles on the network will make roads a safer environment. However, the massive usage of V2X communications leads to an increase of data traffic and resources consumption. Moreover, the mobility associated to vehicles may drain the available resources in particular cells, compromising the required quality of service (QoS) of connected users. To avoid these situations, the introduction of machine learning algorithms to perform mobility load balancing between neighboring cells arises as a promising tool to ensure to all connected vehicles their demanded resources. In this paper, we address the cell overload problem by proposing an O-RAN compliant Q-learning algorithm that dynamically adapts the handover offset between two neighboring cells to mitigate a network overload situation. The algorithm performance is assessed using realistic vehicular traces. Results show that the network overload appearing during rush hour can be successfully mitigated and the load is fairly distributed between cells., This research was partially supported by the Spanish Centre for the Development of Industrial Technology and the Ministry of Economy, Industry and Competitiveness under grant/project CER-20191015 / Open, Virtualized Technology Demonstrators for Smart Networks (Open-VERSO) and by the Spanish Ministry of Economic Affairs and Digital Transformation and the European Union – Next Generation EU, in the framework of the Recovery Plan, Transformation and Resilience (Call UNICO I+D 5G 2021, ref. number TSI063000-2021-29)., Peer Reviewed, Postprint (author's final draft)

With the fast development of Quantum Computers and a better understanding of the quantum realm, Quantum Communications (QC) promises to be one of the main research areas of the upcoming decade. Over the last several years, tremendous effort and progress have already been made in this sense, but until now, QC have primarily been considered at optical frequencies, namely the Near-Infrared (NIR) and visible regions (60-800 THz). This thesis proposes, for the first time, the realization of QC in the terahertz (THz) band (0.1-10 THz), which is expected to become one of the most used bands in the near- immediate future. During the analysis, different possible implementations are considered and discussed. Also, the shape a THz Wireless Quantum channel would take and the main challenges this channel would impose are thoroughly analyzed, considering the wireless channel usually used in quantum optical systems. This thesis is aimed at setting a bench- mark to address the unexplored topic of THz Quantum Communications, enabling future researchers to enter this field and making their work easier., Con el rápido desarrollo de la Computación Cuántica y un mejor conocimiento del mundo cuántico, el campo de las Comunicaciones Cuánticas (QC) promete ser una de las principales áreas de investigación de la próxima década. Durante los últimos años se han realizado considerables esfuerzos y progresos en este campo, pero hasta ahora las QC se han planteado solo para frecuencias ópticas, es decir, las regiones cercanas al infrarojo (NIR) y al visible (60-800 THz). Esta tesis propone, por primera vez, la realización de QC en la banda de terahertz (THz), comprendida entre 0.1 y 10 THz, la cual se espera que sea una de las bandas más utilizadas en el futuro cercano. Durante el análisis se tienen en cuenta y se discuten diferentes posibles implementaciones. También la forma que tendrá el Canal Cuántico Radio a THz y los principales obstáculos que se deberán tener en cuenta, son debidamente analizados partiendo del canal radio considerado en sistemas ópticos. Esta tesis está enfocada en establecer una referencia en el poco explorado campo de las QC a THz, permitiendo a futuros investigadores adentrarse en este campo y facilitarles su trabajo., Amb el ràpid desenvolupament de la Computació Quàntica i un millor coneixement del món quàntic, el camp de les Comunicacions Quàntiques (QC) promet ser una de les principals àrees de recerca de la propera dècada. Durant els darrers anys s'han realitzat considerables esforços i progressos en aquest camp, però fins ara les QC han estat plantejades només per a freqüències òptiques, és a dir, les regions properes a l'infraroig (NIR) i el visible (60-800 THz). Aquesta tesis proposa, per primer cop, la realització de QC a la banda de terahertz (THz), compresa entre 0.1 i 10 THz, la qual s'espera que esdevingui una de les bandes més utilitzades en el futur proper. Durant l'anàlisi es tenen en compte i es discuteixen diferents possibles implementacions. També la forma que obtindria el Canal Quàntic Ràdio a THz, i els principals obstacles que s'haurien de tenir en compte són degudament analitzats, partint del canal ràdio considerat per a sistemes òptics. Aquesta tesis està enfocada a establir una referència en el camp poc explorat de les QC a THz, permetent a futurs investigadors endinsar-se en aquest camp i facilitzar la seva feina.

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)

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)

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 use, 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)

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)

This paper presents a technical overview of BeGREEN project, a Horizon Europe, Smart Networks and Services Joint-Undertaking (SNS-JU) Phase 1 project kicked off on January 1, 2023 [1]. This paper is intended to describe BeGREEN's technical scope and objectives. These objectives aim at improving energy efficiency of the beyond 5G (B5G) networks. BeGREEN technical agenda includes analysis of the combined energy and spectrum efficiency of the B5G networks, based on massive multiple-input-multiple-output (mMIMO) scenarios. The project proposes a novel architecture that includes several innovative solutions. An offloading engine is used for hardware acceleration that is a solution for compute-heavy physical layer processing in 5G new radio (5G NR) mMIMO and beyond to improve the processing performance and energy efficiency. The architecture also includes joint communication and sensing (JCAS) for improving energy efficiency of the physical layer functions by, e.g., efficient beam-search and beam tracking, and uses reconfigurable intelligent surfaces (RIS) as an enabler for JCAS. BeGreenproposes an artificial intelligence (AI)-assisted energy-aware “Intelligent Plane” as an additional plane along with user plane and data plane, that allows the data, model, and inference to be seamlessly exchanged between network functions. The project also proposes an AI Engine that is consist of an execution environment that can host AI models and will manage their lifecycle and access to data., This work is supported by the European Commission’s Horizon Europe, Smart Networks and Services Joint Undertaking, research and innovation program under grant agreement #101097083, BeGREEN project., Peer Reviewed, Postprint (author's final draft)

The capacity sharing problem in Radio Access Network (RAN) slicing deals with the distribution of the capacity available in each RAN node among various RAN slices to satisfy their traffic demands and efficiently use the radio resources. While several capacity sharing algorithmic solutions have been proposed in the literature, their practical implementation still remains as a gap. In this paper, the implementation of a Reinforcement Learning-based capacity sharing algorithm over the O-RAN architecture is discussed, providing insights into the operation of the involved interfaces and the containerization of the solution. Moreover, the description of the testbed implemented to validate the solution is included and some performance and validation results are presented., This paper is part of PORTRAIT project (ref. PDC2021- 120797-I00) funded by MCIN/AEI/10.13039/501100011033 and by European Union Next GenerationEU/PRTR., Peer Reviewed, Postprint (published version)

The increase in complexity of 5G and beyond mobile communications networks to accommodate multiple services with stringent requirements has led to the introduction of Artificial Intelligence (AI) capabilities for automating their management and operation, particularly in the Radio Access Network (RAN). Although there exist a large number of proposals of AI algorithms for different problems in the RAN, little attention has been paid to their practical implementation. This paper intends to fill this gap by discussing the practical aspects on the software development and implementation of AI algorithms for the RAN. This is done based on a specific example that uses deep reinforcement learning for the capacity sharing problem in RAN slicing. The paper presents an implementation of this solution in the context of the O-RAN architecture, detailing the operation of the involved interfaces and the containerization of the solution., Este trabajo es parte del proyecto PORTRAIT (ref. PDC2021-120797-I00) financiado por MCIN/AEI/10.13039/ 501100011033 y por European Union Next Generation EU/PRTR, Peer Reviewed, Postprint (published version)

El aumento de la inteligencia y la capacidad de cálculo de los equipos de usuario (UE) deja la puerta abierta a un nuevo paradigma, en el que los dispositivos personales toman un papel activo en su integración como parte de la RAN. El uso de estas UE como relés es la tendencia dominante, ayudando a otros usuarios al enlace en la estación base cuando experimentan condiciones de mala cobertura. En este contexto, el trabajo desarrollado en este proyecto pretende contribuir y discutir si el UE puede ser consciente de una situación adversa cercana, es decir, pérdida de conexión o handover, estudiando y aplicando técnicas de análisis de series temporales en secuencias reales medidas en células 5G. Los resultados indican que las grabaciones que siguen el mismo camino por lo general tienen suficientes semejanzas para relacionarlos, lo que significa que la historia de un UE anterior se puede utilizar para avanzar una situación que puede empeorar., L'augment de la intel·ligència i la capacitat de càlcul dels equips d'usuari (UE) deixa la porta obert a un nou paradigma, on els dispositius personals prenen un paper actiu en la seva integració com a part de la RAN. L'ús d'aquestes UE com a relés és la tendència dominant, ajudant a altres usuaris a l'enllaç a l'estació base quan experimenten condicions de mala cobertura. En aquest context, el treball desenvolupat en aquest projecte pretén contribuir i discutir si el UE pot ser conscient d'una situació adversa propera, és a dir, pèrdua de connexió o handover, estudiant i aplicant tècniques d'anàlisi de sèries temporals en seqüències reals mesurades en cèl·lules 5G. Els resultats indiquen que els enregistraments que segueixen el mateix camí en general tenen prou semblances per relacionar-los, el que significa que la història de un UE anterior es pot utilitzar per avançar una situació que pot empitjorar., The increased intelligence and computing capacity of user equipment (UE) leaves the door open to a new paradigm, where personal devices take an active role in being integrated as a part of the RAN. The use of these UEs as relays is the dominant trend, assisting other users in the link to the base station when they experience poor coverage conditions. In this context, the work developed in this project is intended to contribute to and discuss if the UE can be aware of an upcoming adverse situation i.e., loss of connection or handover, studying and applying time series analysis techniques in real sequences measured in 5G cells. The findings indicate that recordings following the same path generally have enough similarities to relate them, meaning that previous UE’s history can be used to notice in advance a worsening situation.

This paper envisages a Beyond 5G (B5G) Radio Access Network (RAN) in which the relaying capabilities offered by user equipment (UE) are used as a way to improve the coverage and robustness of the network. The paper proposes and develops the functional framework for supporting the activation of the suitable relay UEs (RUEs) in coverage constrained scenarios. It is based on characterizing each potential RUE through a utility metric that measures the coverage enhancements brought to the network when the RUE is activated. To derive this metric for all the candidate RUEs, the framework considers the use of a Network Digital Twin that allows the offline analysis of different configurations in a fast and safe way. Using the proposed framework, a RUE activation algorithm is proposed and evaluated. The obtained results reflect that significant outage probability reductions can be obtained in the scenario under different traffic distributions thanks to the activation of the RUEs with the highest utility., This paper is part of ARTIST project (ref. PID2020- 115104RB-I00) funded by MCIN/AEI/10.13039/ 501100011033. The work is also funded by the Spanish Ministry of Economic Affairs and Digital Transformation and the European Union - NextGenerationEU under project OPTIMAIX_NDT (Ref. TSI-063000-2021-35)., Peer Reviewed, Postprint (author's final draft)

This paper envisions to augment the Radio Access Network (RAN) infrastructure in Beyond 5G(B5G) systems by exploiting relaying capabilities of user equipment (UE) as a way to improve the coverage, capacity and robustness. Despite the concept and enabling technologies have been in place for some time, their efficient realization requires the conception and development of new features in B5G systems. Among them, this paper focuses on the Relay UE (RUE) activation decision making, in charge of deciding where and when a UE is suitable to be activated to relay traffic from other UEs. Specifically, the paper analyses seven RUE activation strategies that differ on the criteria and the type of context information considered for this decision-making problem. The considered strategies are evaluated through system level simulations in a realistic urban scenario with the objective of assessing the value of each type of context information. Results reveal that the most efficient strategies from the perspective of outage probability reduction are those that account for the number of UEs that would be served by a RUE based on the experienced spectral efficiency., This paper is part of ARTIST project (ref. PID2020-115104RB-I00) funded by MCIN/AEI/10.13039/ 501100011033., Peer Reviewed, Postprint (author's final draft)

The use of multi-connectivity has become a useful tool to manage the traffic in heterogeneous cellular network deployments, since it allows a device to be simultaneously connected to multiple cells. The proper exploitation of this technique requires to adequately configure the traffic sent through each cell depending on the experienced conditions. This motivates this work, which tackles the problem of how to optimally split the traffic among the cells when the multi-connectivity feature is used. To this end, the paper proposes the use of a deep reinforcement learning solution based on a Deep Q-Network (DQN) in order to determine the amount of traffic of a device that needs to be delivered through each cell, making the decision as a function of the current traffic and radio conditions. The obtained results show a near-optimal performance of the DQN-based solution with an average difference of only 3.9% in terms of reward with respect to the optimum strategy. Moreover, the solution clearly outperforms a reference scheme based on Signal to Interference Noise Ratio (SINR) with differences of up to 50% in terms of reward and up to 166% in terms of throughput for certain situations. Overall, the presented results show the promising performance of the DQN-based approach that establishes a basis for further research in the topic of multi-connectivity and for the application of this type of techniques in other problems of the radio access network, This paper is part of ARTIST project (ref. PID2020-115104RB-I00) funded by MCIN/AEI/10.13039/501100011033. The work is also funded by the Spanish Ministry of Science and Innovation under grant ref. PRE2018-084691., Peer Reviewed, Postprint (published version)

Network slicing is a key feature of 5G architecture that allows the partitioning of the network into multiple logical networks, known as network slices, where each of them is customised according to the specific needs of a service or application. Thus, network slicing allows the materialisation of multi-tenant networks, in which a common network infrastructure is shared among multiple communication providers, acting as tenants and each of them using a different network slice. The support of multi-tenancy through slicing in the Radio Access Network (RAN), known as RAN slicing, is particularly challenging because it involves the configuration and operation of multiple and diverse RAN behaviours over the common pool of radio resources available at each of the RAN nodes. Moreover, this configuration needs to be performed in such a way that the specific requirements of each tenant are satisfied and, at the same time, the available radio resources are efficiently used. Therefore, new functionalities that allow the deployment of RAN slices are needed to be introduced at different levels, ranging from Radio Resource Management (RRM) functionalities that incorporate RAN slicing parameters to functionalities that support the lifecycle management of RAN slices. This thesis has addressed this need by proposing, developing and assessing diverse solutions for the support RAN slicing, which has allowed evaluating the capacities, requirements and limitations of network slicing in the RAN from diverse perspectives. Specifically, this thesis is firstly focused on the analytical assessment of RRM functionalities that support multi-tenant and multi-services scenarios, where services are defined according to their 5G QoS requirements. This assessment is conducted through the Markov modelling of admission control policies and the statistical modelling of the resourc allocation, both supporting multiple tenants and multiple services. Secondly, the thesis addresses the problem of slice adm, El Network Slicing és una tecnologia clau de l’arquitectura del 5G que permet dividir la xarxa en múltiples xarxes lògiques, conegudes com a network slices, on cada una es configura d’acord a les necessitats d’un servei o aplicació específic. Així, el network slicing permet la materialització de les xarxes amb múltiples inquilins, on una infraestructura de xarxa comuna es comparteix entre diferents proveïdors de comunicacions, que actuen com a inquilins i utilitzen network slices diferents. El suport de múltiples inquilins mitjançant l’ús del network slicing a la xarxa d’accés ràdio (RAN), que es coneix com a RAN slicing, és un gran repte tecnològic, ja que comporta la configuració i operació de múltiples i diversos comportaments sobre els recursos ràdio disponibles a cadascun dels nodes de la xarxa d’accés. A més a més, aquesta configuració s’ha de portar a terme de forma que els requisits específics de cada inquilí es satisfacin i, al mateix temps, els recursos ràdio disponibles s’utilitzin eficientment. Per tant, és necessari introduir noves funcionalitats a diferents nivells que permetin el desplegament de les RAN slices, des de funcionalitats relacionades amb la gestió dels recursos ràdio (RRM) que incorporin paràmetres per al RAN slicing a funcionalitats que proporcionin suport a la gestió del cicle de vida de les RAN slices. Aquesta tesi ha adreçat aquesta necessitat proposant, desenvolupant i avaluant diverses solucions pel suport del RAN slicing, que han permès analitzar les capacitats, requisits i limitacions del RAN slicing des de diferents perspectives. Específicament, aquesta tesi es centra, en primer lloc, en realitzar una anàlisi de les funcionalitats de RRM que suporten escenaris amb múltiples inquilins i múltiples serveis, on els serveis es defineixen d’acord amb els seus requisits de 5G QoS. Aquesta anàlisi es porta a terme mitjançant la caracterització de polítiques de control d’admissió amb un model de Markov i el modelat estadístic de l’assignac, El Network Slicing es una tecnología clave de la arquitectura del 5G que permite dividir la red en múltiples redes lógicas, conocidas como network slices, que se configuran de acuerdo a las necesidades de servicios y aplicaciones específicas. Así, el network slicing permite la materialización de las redes con múltiples inquilinos, donde una infraestructura de red común se comparte entre diferentes proveedores de comunicaciones, que actúan como inquilinos y que usan network slices diferentes. El soporte de múltiples inquilinos mediante el uso del network slicing en la red de acceso radio (RAN), que se conoce como RAN slicing, es un gran reto tecnológico, ya que comporta la configuración y operación de múltiples y diversos comportamientos sobre los recursos radio disponibles en cada uno de los nodos de la red de acceso. Además, esta configuración debe realizarse de tal manera que los requisitos específicos de cada inquilino se satisfagan y, al mismo tiempo, los recursos radio disponibles se utilicen eficazmente. Por lo tanto, es necesario introducir nuevas funcionalidades a diferentes niveles que permitan el despliegue de las RAN slices, desde funcionalidades relacionadas con la gestión de recursos radio (RRM) que incorporen parámetros para el RAN slicing a funcionalidades que proporcionen soporte a la gestión del ciclo de vida de las RAN slices. Esta tesis ha abordado esta necesidad proponiendo, desarrollando y evaluando diversas soluciones para el soporte del RAN slicing, lo que ha permitido analizar las capacidades, requisitos y limitaciones del RAN slicing desde diversas perspectivas. Específicamente, esta tesis se centra, en primer lugar, en realizar un análisis de funcionalidades de RRM que soportan escenarios con múltiples inquilinos y múltiples servicios, donde los servicios se definen según sus requisitos de 5G QoS. Este análisis se lleva a cabo mediante la caracterización de políticas de control de admisión mediante un modelo de Markov y el modelado a nivel, Postprint (published version)

Multi-connectivity emerges as a useful feature to handle the traffic in heterogeneous cellular scenarios and fulfill the demanding requirements in terms of data rate and reliability. It allows a device to be simultaneously connected to multiple cells belonging to different radio access network nodes from a single or multiple radio access technologies. This paper addresses the problem of optimally splitting the traffic among cells when multi-connectivity is used. For this purpose, it proposes the use of deep learning to determine the optimum amount of traffic of a device that needs to be sent through one or another cell depending on the current traffic and radio conditions. Obtained results reveal a promising capability of the proposed Deep Q Network solution to select quasi optimum traffic splits in the considered scenario., This paper is part of ARTIST project (ref. PID2020- 115104RB-I00) funded by MCIN/AEI/10.13039/ 501100011033. The work is also funded by the Spanish Ministry of Science and Innovation under grant ref. PRE2018-084691., Peer Reviewed, Postprint (published version)

Multi-connectivity, which allows a user equipment to be simultaneously connected to multiple cells from different radio access network nodes that can be from a single or multiple radio access technologies, has emerged as a useful feature to handle the traffic in heterogeneous cellular scenarios and fulfill high data rate and reliability requirements. This paper proposes the use of deep reinforcement learning to optimally split the traffic among cells when multi-connectivity is considered in a heterogeneous 4G/5G networks scenario. Obtained results reveal a promising capability of the proposed Deep Q Network solution to select quasi optimum traffic splits depending on the current traffic and radio conditions in the considered scenario. Moreover, the paper analyses the robustness of the obtained policy in front of variations with respect to the conditions used during the training., This paper is part of ARTIST project (ref. PID2020-115104RB-I00) funded by MCIN/AEI/10.13039/ 501100011033. The work is also funded by the Spanish Ministry of Science and Innovation under grant ref. PRE2018-084691., Peer Reviewed, Postprint (published version)

Reinforcement Learning (RL)-based algorithmic solutions have been profusely proposed in recent years for addressing multiple problems in the Radio Access Network (RAN). However, how RL algorithms have to be trained for a successful exploitation has not received sufficient attention. To address this limitation, which is particularly relevant given the peculiarities of wireless communications, this paper proposes a functional framework for training RL strategies in the RAN. The framework is aligned with the O-RAN Alliance machine learning workflow and introduces specific functionalities for RL, such as the way of specifying the training datasets, the mechanisms to monitor the performance of the trained policies during inference in the real network, and the capability to conduct a retraining if necessary. The proposed framework is illustrated with a relevant use case in 5G, namely RAN slicing, by considering a Deep Q-Network algorithm for capacity sharing. Finally, insights on other possible applicability examples of the proposed framework are provided. © 2022 IEEE., This paper is part of ARTIST project (ref. PID2020-115104RB-I00) funded by MCIN/AEI/10.13039/ 501100011033 and PORTRAIT project (ref. PDC2021-120797-I00) funded by MCIN/AEI/10.13039/501100011033 and by European Union Next GenerationEU/PRTR., Postprint (published version)

Fifth Generation (5G) networks will support countless new applications and new business models. One of the 5G paradigms is network slicing, which enables the integration of multiple logical networks each one tailored to the requirements of the different services that can be provided by both network operators and vertical industries. One of the services where 5G is expected to have a greatest impact is vehicular-to-everything (V2X) communications, which will have their stringent latency requirements now met. However, the mobility associated to vehicles can lead to cell overload compromising the required quality of service (QoS). To address this problem, in this paper we propose and evaluate the performance of three network overload alleviation techniques to control network congestion provoked by traffic jams using realistic vehicular traces in a network slicing environment. Firstly, we describe the architecture supporting V2X communications. Secondly, the network congestion control approaches are explained. Finally, after providing a complete description of the considered scenario, results will be detailed, showing that the network overload appearing during rush hour can be significantly reduced., This research was supported by the Spanish Centre for the Development of Industrial Technology (CDTI) and the Ministry of Economy, Industry and Competitiveness under grant/project CER-20191015 / Open, Virtualized Technology Demonstrators for Smart Networks (Open-VERSO)., Peer Reviewed, Postprint (author's final draft)

Deep Reinforcement Learning (DRL) has recently emerged as a promising technique to deal with different problems in the 5G and beyond Radio Access Network (RAN). The practical implementation of DRL solutions in the real network embraces a training process that is fundamental to materialize the expected benefits associated to these techniques. However, little effort has been devoted in the literature to analyze this training process when applying DRL in the RAN. In an effort to contribute to fill this gap, this paper presents an impact analysis of the training process on the obtained performance by a DRL solution for RAN slicing. To this end, a methodology to specify the training dataset is introduced together with the definition of relevant metrics. Then, the paper presents different simulation results to determine the features of the training dataset that allow a satisfactory training and a high performance of the obtained policy when applied during the inference stage., Peer Reviewed, Postprint (published version)

Next generation of mobile communications will have to cover the unceasingly growing demand of higher data rates and reliable services. In this situation it will be a must to migrate towards higher frequencies in the range of millimeter waves (mm-Waves), where there is plenty of available bandwidth, that would allow high data rates. However, for this new range of the spectrum the attenuation suffered is very strong. Therefore, the isotropic antennas used for lower frequencies are going to be replaced for highly directive ones in order to compensate for this effect. But at the same time the use of narrow beam will introduce a new problematic aspect that was previously inexistent, which is beam selection and beam alignment. In order to address this situation, efficient beamforming and beam alignment becomes critical. The ultimate aim of this thesis is to test Frequency Scanning Arrays (FSA) for analog beam probing techniques in multi-user hybrid beamforming systems. In order to achieve it, several previous steps have been made. This thesis is aiming to apply the technology of Frequency Scanning Arrays (FSA) in multiuser hybrid beamforming systems. First the analog beam probing using traditional Exhaustive Search (ES) is implemented, with not only hybrid but also analog and digital schemes. Also, several beam selection algorithms are defined and tested. Afterwards the effect of the channel estimation is considered and a timing delay for each user is implemented, in order to make the final comparison between ES and FSA as realistic as possible. The results obtained show that in most cases analised the FSA system outperforms the ES system. These encouraging results could be a good indicator that the use of FSA analog beam probing has many promising possibilities and has to be further investigated., La próxima generación de comunicaciones móviles tendrá que cubrir la incesantemente creciente demanda de mayores velocidades de transmisión datos y servicios fiables. En esta situación será imprescindible migrar hacia frecuencias más altas en el rango de las ondas milimétricas (mm-Waves), donde hay mucho ancho de banda disponible, que permitiría altas velocidades de transferencia de datos. Sin embargo, para este nuevo rango del espectro la atenuación sufrida es muy fuerte. Por ello, las antenas isotrópicas utilizadas para frecuencias más bajas van a ser sustituidas por otras muy directivas para compensar este efecto. Pero al mismo tiempo, el uso de un haz estrecho introducirá una nueva problemática que antes no existía, que es la selección y la alineación del haz. Para hacer frente a esta situación, una eficiente formación y alineación del haz se convierte en un aspecto fundamental. El objetivo final de esta tesis es probar los Frequency Scanning Arrays (FSA) para las técnicas de sondeo del haz analógico en sistemas de conformación de haz híbrido multiusuario. Para conseguirlo, se han realizado varios pasos previos. En esencia, en esta tesis se propone aplicar la tecnología de los Frequency Scanning Arrays (FSA) en sistemas multiusuario. En primer lugar, se implementa el sondeo analógico del haz mediante la tradicional búsqueda exhaustiva (Exhaustive Search, ES) con esquemas no sólo híbridos sino también analógicos y digitales. Además, se definen y prueban varios algoritmos de selección de haz. Después se considera el efecto de la estimación del canal y se implementa un retardo temporal para cada usuario, con el fin de realizar la comparación final entre ES y FSA de la manera más realista posible. Los resultados obtenidos muestran que en la mayoría de los casos analizados el sistema FSA supera al sistema ES. Estos resultados alentadores podrían ser un buen indicador de que el uso del haz analógico FSA tiene prometedoras posibilidades y tiene que ser investigado más, La pròxima generació de comunicacions mòbils haurà de cobrir la incessant i creixent demanda d'altes velocitats de transmissió de dades i serveis cada cop més fiables. En aquesta situació serà imprescindible migrar cap a freqüències més altes en el rang de les ones mil·limètriques (mm-Waves), on hi ha un gran ample de banda disponible, que permetria altes velocitats de transferència de dades. Tot i això, per aquest nou rang de l'espectre, l'atenuació que pateix el senyal és molt gran. Per aquesta raó, les antenes isotròpiques utilitzades per a freqüències més baixes seran substituïdes per d'altres molt directives per a compensar aquest efecte. Però al mateix temps, l'ús d'un feix estret introduirà una nova problemàtica que abans no existia, que és la selecció i alineació del feix. Per fer front a aquesta situació, una formació i alineació eficient del feix esdevé un aspecte fonamental. L'objectiu final d'aquesta tesi és provar els Frequency Scanning Arrays (FSA) per a les tècniques de sondeig del feix analògic en sistemes de conformació de feix híbrid multiusuari. Per tal d'aconseguir-ho s'han dut a terme diversos passos previs. En essència, en aquesta tesi es proposa aplicar la tecnologia dels Frequency Scanning Arrays (FSA) en sistemes multiusuari. En primer lloc, s'implementa el sondeig analògic del feix mitjançant la cerca exhaustiva tradicional (Exhaustive Search, ES) amb esquemes no només híbrids sinó també analògics i digitals. A més a més, es defineixen i es proven diversos algoritmes de selecció de feix. Després es considera l'efecte de l'estimació del canal i s'implementa un retard temporal per a cada usuari, amb la finalitat de dur a terme la comparació final entre ES i FSA de la manera més realista possible. Els resultats obtinguts demostren que en la majoria dels casos analitzats el sistema FSA supera al sistema ES. Aquests resultats encoratjadors podrien ser un bon indicador que l'ús del feix analògic FSA ofereix possibilitats prometedores i ha de ser

One of the key features of the 5G architecture is network slicing, which allows the simultaneous support of diverse service types with heterogeneous requirements over a common network infrastructure. In order to support this feature in the Radio Access Network (RAN), it is required to have capacity sharing mechanisms that distribute the available capacity in each cell among the existing RAN slices while satisfying their requirements and efficiently using the available resources. Deep Reinforcement Learning (DRL) techniques are good candidates to deal with the complexity of capacity sharing in multi-cell scenarios where the traffic in the different cells can be heterogeneously distributed in the time and space domains. In this paper, a multi-agent reinforcement learning-based solution for capacity sharing in multi-cell scenarios is discussed and assessed under heterogeneous traffic conditions. Results show the capability of the solution to satisfy the requirements of the RAN slices while using the resources in the different cells efficiently., This work has been supported by the Spanish Research Council and FEDER funds under SONAR 5G grant (ref.TEC2017-82651-R), by the European Commission’s Horizon 2020 5G-CLARITY project under grant agreement 871428 and by the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia under grant 2020FI_B2 00075., Peer Reviewed, Postprint (author's final draft)

Along the successive generations, mobile network infrastructure deployments have required continuous expansion and upgrades to respond to the ever increasing traffic demand. At the same time, user equipment (UE) has evolved with the introduction of powerful and advanced communication and computational capabilities. Moreover, several industrial initiatives are promoting the principles of RAN disaggregation through the split of RAN functionalities among multiple interoperable components in contrast to traditional monolithic solutions. Embracing these trends, this paper presents the vision of a Beyond 5G (B5G) scenario where the UE takes a more active role in the communication service provisioning and is used to augment the RAN infrastructure as a source of distributed capacity and network intelligence. This is expected to bring benefits for Mobile Network Operators by reducing the number of base stations to deploy and by providing the RAN with a new degree of flexibility for smartly and dynamically adapting the RAN to the specific needs. In this direction, after discussing the main enabling technologies of the envisaged RAN augmentation concept, this paper presents a simulation-based assessment to quantify the performance improvements and the achievable savings in terms of infrastructure deployment., This paper is part of ARTIST project (ref. PID2020- 115104RB-I00) funded by MCIN/AEI/10.13039/ 501100011033., Peer Reviewed, Postprint (author's final draft)

© 2021 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., 5G is envisioned to simultaneously provide diverse service types with heterogeneous needs under very different application scenarios and business models. Therefore, network slicing is included as a key feature of the 5G architecture to allow sharing a common infrastructure among different tenants, such as mobile communication providers, vertical market players, etc. In order to provide the Radio Access Network (RAN) with network slicing capabilities, mechanisms that efficiently distribute the available capacity among the different tenants while satisfying their needs are required. For this purpose, this paper proposes a multi-agent reinforcement learning approach for RAN capacity sharing. It makes use of the Deep Q-Network algorithm in a way that each agent is associated to a different tenant and learns the capacity to be provided to this tenant in each cell while ensuring that the service level agreements are satisfied and that the available radio resources are efficiently used. The consideration of multiple agents contributes to a better scalability and higher learning speed in comparison to single-agent approaches. In this respect, results show that the policy learnt by the agent of one tenant can be generalised and directly applied by other agents, thus reducing the complexity of the training and making the proposed solution easily scalable, e.g., to add new tenants in the system. The proposed approach is well aligned with the on-going 3GPP standardization work and guidelines for the parametrization of the solution are provided, thus enforcing its practical applicability., This work was supported in part by the Spanish Research Council and FEDER funds under SONAR 5G Grant ref. TEC2017-82651-R, in part by the European Commission’s Horizon 2020 5G-CLARITY project under Grant 871428, and in part by the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia under Grant 2020FI_B2 00075., Peer Reviewed, Postprint (author's final draft)

Network slicing is included as a key feature of the 5G architecture in order to simultaneously support diverse service types with heterogeneous requirements. The realization of network slicing in the Radio Access Network (RAN) needs mechanisms that allow the distribution of the available capacity in the system in an efficient manner while satisfying the requirements of the different services. In this paper, a capacity sharing function is proposed, which is approached as a multi agent reinforcement learning based on the Deep Reinforcement Learning (DRL) algorithm Deep Q-Network (DQN). The proposed algorithm provides the capacity to be assigned to each RAN slice. Performance assessment reveals the promising behavior of the proposed solution., This work has been supported by the Spanish Research Council and FEDER funds under SONAR 5G grant (ref. TEC2017-82651-R), by the European Commission’s Horizon 2020 research and innovation program under grant agreement #871428, 5G-CLARITY project, and by the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia under grant 2019FI_B1 00102., Peer Reviewed, Postprint (author's final draft)

© 2020 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., Along with supporting new business opportunities, data analytics –understood as the pursuit of extracting meaning from raw data– will be key to getting 5G Radio Access Networks (RANs) efficiently rolled out and operated. The exploitation of data analytics services within the RAN itself, that is, to empower the operation of the RRM algorithms executed within the RAN nodes, is an incipient area of interest that broadens the more well-established applicability of using data analytics for on-top-of-the-RAN operations. In this respect, this article is an attempt to put a spotlight on this topic by motivating a plausible data analytics architectural framework for data-driven, smarter RRM algorithms., This work has been supported by the Spanish Research Council and FEDER funds under SONAR 5G grant (ref. TEC2017-82651-R)., Peer Reviewed, Postprint (author's final draft)

Network slicing is a central feature in 5G and beyond systems to allow operators to customize their networks for different applications and customers. With network slicing, different logical networks, i.e. network slices, with specific functional and performance requirements can be created over the same physical network. A key challenge associated with the exploitation of the network slicing feature is how to efficiently allocate underlying network resources, especially radio resources, to cope with the spatio-temporal traffic variability while ensuring that network slices can be provisioned and assured within the boundaries of Service Level Agreements / Service Level Specifications (SLAs/SLSs) with customers. In this field, the use of artificial intelligence, and, specifically, Machine Learning (ML) techniques, has arisen as a promising approach to cater for the complexity of resource allocation optimization among network slices. This paper tackles the description of a feasible implementation framework for deploying ML-assisted solutions for cross-slice radio resource optimization that builds upon the work conducted by 3GPP and O-RAN Alliance. On this basis, the paper also describes and evaluates an ML-assisted solution that uses a Multi-Agent Reinforcement Learning (MARL) approach based on the Deep Q-Network (DQN) technique and fits within the presented implementation framework., This work has been supported by the Spanish Research Council and FEDER funds under SONAR 5G grant (ref. TEC2017-82651-R) and by the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia under grant 2019FI_B1 00102., Peer Reviewed, Postprint (published version)

5G systems are envisaged to support a wide range of application scenarios with variate requirements. To handle this heterogeneity, 5G architecture includes network slicing capabilities that facilitate the partitioning of a single network infrastructure into multiple logical networks on top of it, each tailored to a given use case and provided with appropriate isolation and Quality of Service (QoS) characteristics. Network slicing also enables the use of multi-tenancy networks, in which the same infrastructure can be shared by multiple tenants by associating one slice to each tenant, easing the cost-effective deployment and operation of future 5G networks. Concerning the Radio Access Network (RAN), slicing is particularly challenging as it implies the configuration of multiple RAN behaviors over a common pool of radio resources. In this context, this work presents a Markov model for RAN slicing capable of characterizing diverse Radio Resource Management (RRM) strategies in multi-tenant and multi-service 5G scenarios including both guaranteed and non-guaranteed bit rate services. The proposed model captures the fact that different radio links from diverse users can experience distinct spectral efficiencies, which enables an accurate modeling of the randomness associated with the actual resource requirements. The model is evaluated in a multi-tenant scenario in urban micro cell and rural macro cell environments to illustrate the impact of the considered RRM polices in the QoS provisioning., Peer Reviewed, Postprint (published version)

Network slicing has become a fundamental capability for 5G networks to support the expected high variety of service requirements over a common physical network infrastructure. Each network slice can be customized for a specific application, making that the radio resources have to be accordingly managed by the Radio Access Network (RAN) part of the slice. In this thesis, three different Deep Reinforcement Learning (DRL) based approaches are presented to optimize the resource allocation among slices. A RAN slicing simulator scenario is developed, where the DRL mechanisms build knowledge about the network and learn how to optimize the capacity allocation for each tenant at every moment of time. The performance of each approach is studied based on simulation results, and before the comparison between the algorithms, the set of hyperparameters of each approach is tuned to optimize the learning process.

The new 5th generation (5G) era has transformed previous mobile generations into fast, smart networks that will be more responsive and customizable. With network slicing, 5G networks can be dynamically adapted to the different needs of specific vertical industries. This capability has opened the opportunity to new business models whereby infrastructure owners can monetize their investment by leasing resources to third parties (i.e. tenants). In this respect, a challenging task for the owner of the radio access network infrastructure (i.e. the network provider) is the spectrum planning of multi-tenant scenarios. This paper proposes different alternatives of hiring capacity to the provider as well as a set of spectrum planning strategies, each giving a certain degree of flexibility to allocate resources per tenant. These strategies are evaluated in a 5G small cell multi-tenant network through snapshot-based simulations. The performance of the strategies is assessed in terms of scalability, spectrum isolation, utilization and efficiency., This work was supported in part by the H2020 Research and Innovation Project 5G-CLARITY under Grant 871428, in part by theAndalusian Knowledge Agency under Project A-TIC-241-UGR18, in part by the Spanish Ministry of Education, Culture and Sport (FPU)under Grant 17/01844, and in part by the Spanish Research Council and FEDER funds through SONAR 5G under GrantTEC2017-82651-R, Peer Reviewed, Postprint (published version)

© 2019 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., 2019 Best Paper Runner up Award, atorgat per The 6th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM)., While the support of Mission Critical (MC) communications on commercial cellular networks has already been incorporated in the latest releases of Long Term Evolution (LTE), it is expected that the network slicing feature of Fifth Generation (5G) systems will further boost the provision of these services thanks to the possibility of creating customized and isolated network slices adapted to the specific requirements of MC communications. At the Radio Access Network (RAN), the realization of a network slice requires to specify how the pool of available radio resources is split between the different slices in accordance with their service requirements. In this context, this paper addresses the use of RAN slicing for provisioning MC services taking as a reference the emergency scenario defined by the 5G ESSENCE project. It is characterized by different stages associated to the occurrence of an incident and its evolution, thus involving different communication needs. For each stage, an estimation of the capacity requirements to be granted to the MC RAN slice is provided. Then, the architecture of the project is discussed, focusing on the components that enable the RAN slicing management to properly support MC services., Peer Reviewed, Award-winning, Postprint (author's final draft)

This is the peer reviewed version of the following article: Mendoza, F, Ferrus, R, Sallent, O. Experimental proof of concept of an SDN‐based traffic engineering solution for hybrid satellite‐terrestrial mobile backhauling. Int J Satell Commun Network. 2019; 37: 630– 645, which has been published in final form at https://doi.org/10.1002/sat.1303. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving, Satellite networks are expected to be an integral part of 5G service deployment. One compelling use case is mobile backhauling, where the exploitation of a satellite component can improve the reach, robustness, and economics of 5G rollout. The envisaged availability of new satellite capacity, together with the development of better integration approaches for the provisioning and operation of the satellite component in a more flexible, agile, and cost-effective manner than done today, are expected to revamp such use case within the 5G ecosystem. In this context, sustained in the architectural designs proposed within H2020 VITAL research project, this paper presents an experimental proof of concept (PoC) of a satellite-terrestrial integration solution that builds upon software-defined networking (SDN) technologies for the realization of end-to-end traffic engineering (E2E TE) in mobile backhauling networks with a satellite component. A laboratory test bed has been developed and validated, consisting of a small-scale private mobile network with a backhaul setting that combines Ethernet-wired links, a satellite link emulator (OpenSAND), OpenFlow switches, and an OpenFlow controller running the network application for E2E TE. Provided results show the operation of a E2E TE application able to enforce different traffic routing and path failure restoration policies as well as the performance impact that it has on the mobile network connectivity services., Peer Reviewed, Postprint (author's final draft)

Network slicing is a fundamental feature of 5G systems that allows the partitioning of a single network into a number of segregated logical networks, each optimized for a particular type of service, or dedicated to a particular customer or application. While support for network slicing (e.g. identifiers, functions, signalling) is already defined in the latest 3GPP Release 15 specifications, solutions for efficient automated management of network slicing (e.g. automatic provisioning of slices) are still at a much more incipient stage, especially for what concerns the next-generation Radio Access Network (NG-RAN). In this context, and consistently with the new service-based management architecture defined by 3GPP for 5G systems, this paper presents a functional framework for the management of network slicing in a NG-RAN infrastructure, delineating the interfaces and information models necessary to support the dynamic and automatic deployment of RAN slices. A discussion on the complexity of such automation follows together with an illustrative description of the applicability of the overall framework and information models in the context of a neutral host provider scenario that offers RAN slices to third party service providers., Peer Reviewed, Postprint (published version)

Radio Measurements in LTE: Collection and Analysis, On the 4G networks and also on the future 5G is wished to put forward an auto-management of the network in a way that certain functionalities, such as its optimization, can be carried out in a more automatic way. The project has been developed with a test LTE base station and a group of measurement terminals, which have allowed to have an approach to the issues that the whole process involve. However, on the other hand, ideas have been proposed about the expectations of these type of future systems. The tests have been done by means of the recreation of potential scenes in which the applications of the system could improve their behaviour thanks to the measurements. The improvements could be, for instance, on the fields of the Minimization of Drive Tests, the detection of excessive interference, readjustment of the scheduler or the upgrade of the handover processes., En las redes 4G y en el futuro 5G se quiere avanzar hacia una autogestión de la red de manera que ciertas funcionalidades, como su optimización, se puedan llevar a cabo de una forma más automática. El proyecto se ha desarrollado con una estación base LTE de pruebas y una serie de terminales de medidas con las que se ha permitido tener un enfoque sobre las problemáticas que todo este procesado de datos comporta, pero, por otro lado, proponiendo ideas de lo que se espera de este tipo de sistemas en el futuro. Estas pruebas se han hecho mediante la recreación de escenarios potenciales donde las aplicaciones de este sistema podrían mejorar su funcionamiento gracias a esta toma de medidas, como podrían ser la minimización de Drive Tests, la detección de interferencia excesiva, el reajuste del organismo que asigna los recursos radio o la mejora de los procesos de cambio de celda., A les xarxes 4G i en el futur al 5G es vol avançar cap a una autogestió de la xarxa de manera que certes funcionalitats, com la seva optimització, es puguin dur a terme d'una forma més automàtica. El projecte s'ha desenvolupat amb una estació base LTE de proves i una sèrie de terminals de mesures amb els que s'ha permès tenir un enfoc sobre les problemàtiques que tot aquest processament de dades comporta però, d'altra banda, proposant idees del que s'espera d'aquest tipus de sistemes en el futur. Aquestes proves s'han fet mitjançant la recreació de potencials escenaris on les aplicacions d'aquest sistema podrien millorar el seu funcionament gràcies a aquesta presa de mesures, com podrien ser la minimització de Drive Tests, la detecció d'interferència excessiva, el reajustament de l'organisme que assigna els recursos ràdio o la millora dels processos de canvis de cel·la.