The recent years have witnessed an unprecedented growth in the number of connected devices and amount of bandwidth required by the multiple services offered by wireless devices. The current 5G standard addresses such issues by adopting higher carrier frequencies and antennas with a large number of radiating elements. The former solution enables to exploit larger bandwidths in the millimeter-wave (mmWave) portion of the spectrum, while the latter one allows access points to serve an increasingly higher number of users. Both find realization in the Multiple-Input-Multiple-Output (MIMO) antenna systems with their enhanced beamforming capabilities. While the adoption of the hybrid digital-analog beamforming architecture lightens the overall system complexity, the need of miniaturized, high-performance and broadband hardware components is still an open issue. Passive Radio Frequency (RF) components in MicroElectroMechanical-Systems technology (RF-MEMS) offer notable and broadband electrical performances, while maintaining the marked miniaturization required for the hardware to be employed in the MIMO antennas, characterizing the current and future telecommunications scenario. Whilst numerous examples of single RF-MEMS switches, attenuators and phase shifters are available in the literature since about two decades, still limited attention is dedicated to the development of MEMS-based multi-device monolithic networks embedding such devices. High-performance RF-MEMS networks of this kind could represent the base of future MIMO beamforming architectures. Given such a context, the fundamental core of this thesis is the design and the realization of ad hoc RF-MEMS devices to be integrated in a reconfigurable monolithic module, operating in the realistic scenario of the mm-Wave portion of the spectrum allocated to 5G in Europe (24.25–27.5 GHz). The resulting devices consist in a 3-bit attenuator, three 1-bit phase-shifting cells and a Single-Pole-Double-Throw (SPDT) switch, each relying on membranes featuring a reduced actuation voltage, in the 5–9 V range, for an easier interfacing with electronics based on Complementary Metal–Oxide–Semiconductor (CMOS). To this purpose, the ad hoc designed MEMS switching membranes, along with prototypes of the building blocks to be embedded in the final module, are designed, optimized and fabricated. The experimental measurements performed on the prototypes of membranes (i.e. micro-switches), attenuation cells, optimized resistors and a phase shifter are compared to FEM-based (Finite Element Method) simulated results. Such comparison validates the simulation approach, in both the electromagnetic and the electro-mechanical domains, by which the proposed module is then designed and optimized in its final layout. To the best of our knowledge, this project is among the first to investigate the development of a monolithic module, entirely based on RF-MEMS passives, implementing both the attenuation and the phase shifting functionalities that can be employed in hybrid beamforming architectures at each antenna element. More in detail, the module features at least 25 attenuation and phase shifting states, from -5.39 dB to -13.51 dB by variable steps, and from 10.59° to 158.46°, respectively. Concerning the SPDT switch, satisfying electrical performances have been demonstrated in terms of return loss (<-10 dB), insertion loss (<-1.2 dB) and isolation (<-25 dB) over the 0–30 GHz interval. Despite their increased complexity, appealing results have marked the proposed attenuator and the phase-shifting cells, whose return and insertion losses are always better than -10 dB and -3 dB, respectively, along the frequency interval of interest. With an overall footprint not exceeding 9.51x3.35 mm2, the designed module effectively combines the miniaturization, broadband, and linear electrical behavior of RF-MEMS, making it a suitable candidate for the MIMO antennas of the current and future telecommunications scenario.

Yes / As 5th Generation research reaches the twilight, the research community must go beyond 5G and look towards the 2030 connectivity landscape, namely 6G. In this context, this work takes a step towards the 6G vision by proposing a next generation communication platform, which aims to extend the rigid coverage area of fixed deployment networks by considering virtual mobile small cells (MSC) that are created on demand. Relying on emerging computing paradigms such as NFV (Network Function Virtualization) and SDN (Software Defined Networking), these cells can harness radio and networking capability locally reducing protocol signalling latency and overhead. These MSCs constitute an intelligent pool of networking resources that can collaborate to form a wireless network of MSCs providing a communication platform for localized, ubiquitous and reliable connectivity. The technology enablers for implementing the MSC concept are also addressed in terms of virtualization, lightweight wireless security, and energy efficient RF. The benefits of the MSC architecture towards reliable and efficient cell-offloading are demonstrated as a use-case. / This project has received funding from the European Union's H2020 research and innovation program under grant agreement H2020-MCSAITN- 2016-SECRET 722424 [2].

Yes / The world is experiencing an explosion in demand for ultra-high data rates with far greater expectations in the next few years. These expectations, given the bandwidth-demanding applications such as augmented and virtual reality and other beyond-5G applications, motivate the exploration of higher-frequency communication in the terahertz (THz) bands. However, THz communication is faced with many technical challenges, primarily due to the high susceptibility to blockages that limit its applications. Here, reconfigurable intelligent surfaces (RIS) provide alternative paths to circumvent such blockage effects and ensure reliable, spectral, and energy-efficient communication, thus advancing the THz-RIS technology concept. However, the ambitious targets of ubiquitous and global connectivity can only be satisfied by many technologies extending to multiple domains, from terrestrial networks to non-terrestrial network (NTN) domains. The use of airborne and spaceborne networks is considered a potential solution for addressing these challenges due to their dynamism, coverage, and ability to leverage their altitude for achieving line-of-sight communication for enhanced signal quality and network performance. Therefore, unmanned aerial vehicles, high-altitude platform stations, and satellites are poised to use flying THz-based RISs to improve air-to-ground and space-to-ground communication reliability while exploiting novel RIS architectures, techniques and enablers to address the issues regarding the propagation conditions, hardware limitations, network complexity and system performance. The aim in this paper is to present the discussion and a survey on the technical advances on THz-RIS NTNs, in addition to outlining potential applications, architectural variants, influencing properties, as well as its prospects, associated challenges, open issues and future directions towards high-data rate THz-RIS NTN communication for 6G and beyond. / This work was supported in part by the Universiti Kebangsaan Malaysia through Dana Impak Perdana 2.0 under Grant DIP-2022-020; and in part by the Engineering and Physical Sciences Research Council [grant number EP/Z001544/1] through the UK Research and Innovation (UKRI)-funded Marie Skłodowska-Curie Actions (MSCA) Postdoctoral Fellowship

Tesis por compendio / [ES] La proliferación de sistemas autónomos y su creciente integración en la vida humana cotidiana han abierto nuevas fronteras de investigación y desarrollo. Dentro de este ámbito, la presente tesis se adentra en las aplicaciones multifacéticas de los LLMs (Large Language Models), técnicas de DL (Deep Learning) y algoritmos de optimización en el ámbito de estos sistemas autónomos. A partir de los principios de los métodos potenciados por la Inteligencia Artificial (IA), los estudios englobados en este trabajo convergen en la exploración y mejora de distintos sistemas autónomos que van desde sistemas de platooning de camiones en sistemas de comunicaciones Beyond 5G (B5G), Sistemas Multi-Agente (SMA), Vehículos Aéreos No Tripulados (UAV), estimación del área de incendios forestales, hasta la detección temprana de enfermedades como el glaucoma. Un enfoque de investigación clave, perseguido en este trabajo, gira en torno a la implementación innovadora de controladores PID adaptativos en el platooning de vehículos, facilitada a través de la integración de los LLMs. Estos controladores PID, cuando se infunden con capacidades de IA, ofrecen nuevas posibilidades en términos de eficiencia, fiabilidad y seguridad de los sistemas de platooning. Desarrollamos un modelo de DL que emula un controlador PID adaptativo, mostrando así su potencial en las redes y radios habilitadas para IA. Simultáneamente, nuestra exploración se extiende a los sistemas multi-agente, proponiendo una Teoría Coevolutiva Extendida (TCE) que amalgama elementos de la dinámica coevolutiva, el aprendizaje adaptativo y las recomendaciones de estrategias basadas en LLMs. Esto permite una comprensión más matizada y dinámica de las interacciones estratégicas entre agentes heterogéneos en los SMA. Además, nos adentramos en el ámbito de los vehículos aéreos no tripulados (UAVs), proponiendo un sistema para la comprensión de vídeos que crea una log de la historia basada en la descripción semántica de eventos y objetos presentes en una escena capturada por un UAV. El uso de los LLMs aquí permite razonamientos complejos como la predicción de eventos con mínima intervención humana. Además, se aplica una metodología alternativa de DL para la estimación del área afectada durante los incendios forestales. Este enfoque aprovecha una nueva arquitectura llamada TabNet, integrada con Transformers, proporcionando así una estimación precisa y eficiente del área. En el campo de la salud, nuestra investigación esboza una metodología exitosa de detección temprana del glaucoma. Utilizando un enfoque de entrenamiento de tres etapas con EfficientNet en imágenes de retina, logramos una alta precisión en la detección de los primeros signos de esta enfermedad. A través de estas diversas aplicaciones, el foco central sigue siendo la exploración de metodologías avanzadas de IA dentro de los sistemas autónomos. Los estudios dentro de esta tesis buscan demostrar el poder y el potencial de las técnicas potenciadas por la IA para abordar problemas complejos dentro de estos sistemas. Estas investigaciones en profundidad, análisis experimentales y soluciones desarrolladas arrojan luz sobre el potencial transformador de las metodologías de IA en la mejora de la eficiencia, fiabilidad y seguridad de los sistemas autónomos, contribuyendo en última instancia a la futura investigación y desarrollo en este amplio campo. / [CA] La proliferació de sistemes autònoms i la seua creixent integració en la vida humana quotidiana han obert noves fronteres de recerca i desenvolupament. Dins d'aquest àmbit, la present tesi s'endinsa en les aplicacions multifacètiques dels LLMs (Large Language Models), tècniques de DL (Deep Learning) i algoritmes d'optimització en l'àmbit d'aquests sistemes autònoms. A partir dels principis dels mètodes potenciats per la Intel·ligència Artificial (IA), els estudis englobats en aquest treball convergeixen en l'exploració i millora de diferents sistemes autònoms que van des de sistemes de platooning de camions en sistemes de comunicacions Beyond 5G (B5G), Sistemes Multi-Agent (SMA), Vehicles Aeris No Tripulats (UAV), estimació de l'àrea d'incendis forestals, fins a la detecció precoç de malalties com el glaucoma. Un enfocament de recerca clau, perseguit en aquest treball, gira entorn de la implementació innovadora de controladors PID adaptatius en el platooning de vehicles, facilitada a través de la integració dels LLMs. Aquests controladors PID, quan s'infonen amb capacitats d'IA, ofereixen noves possibilitats en termes d'eficiència, fiabilitat i seguretat dels sistemes de platooning. Desenvolupem un model de DL que emula un controlador PID adaptatiu, mostrant així el seu potencial en les xarxes i ràdios habilitades per a IA. Simultàniament, la nostra exploració s'estén als sistemes multi-agent, proposant una Teoria Coevolutiva Estesa (TCE) que amalgama elements de la dinàmica coevolutiva, l'aprenentatge adaptatiu i les recomanacions d'estratègies basades en LLMs. Això permet una comprensió més matissada i dinàmica de les interaccions estratègiques entre agents heterogenis en els SMA. A més, ens endinsem en l'àmbit dels Vehicles Aeris No Tripulats (UAVs), proposant un sistema per a la comprensió de vídeos que crea un registre de la història basat en la descripció semàntica d'esdeveniments i objectes presents en una escena capturada per un UAV. L'ús dels LLMs aquí permet raonaments complexos com la predicció d'esdeveniments amb mínima intervenció humana. A més, s'aplica una metodologia alternativa de DL per a l'estimació de l'àrea afectada durant els incendis forestals. Aquest enfocament aprofita una nova arquitectura anomenada TabNet, integrada amb Transformers, proporcionant així una estimació precisa i eficient de l'àrea. En el camp de la salut, la nostra recerca esbossa una metodologia exitosa de detecció precoç del glaucoma. Utilitzant un enfocament d'entrenament de tres etapes amb EfficientNet en imatges de retina, aconseguim una alta precisió en la detecció dels primers signes d'aquesta malaltia. A través d'aquestes diverses aplicacions, el focus central continua sent l'exploració de metodologies avançades d'IA dins dels sistemes autònoms. Els estudis dins d'aquesta tesi busquen demostrar el poder i el potencial de les tècniques potenciades per la IA per a abordar problemes complexos dins d'aquests sistemes. Aquestes investigacions en profunditat, anàlisis experimentals i solucions desenvolupades llançen llum sobre el potencial transformador de les metodologies d'IA en la millora de l'eficiència, fiabilitat i seguretat dels sistemes autònoms, contribuint en última instància a la futura recerca i desenvolupament en aquest ampli camp. / [EN] The proliferation of autonomous systems, and their increasing integration with day-to-day human life, have opened new frontiers of research and development. Within this scope, the current thesis dives into the multifaceted applications of Large Language Models (LLMs), Deep Learning (DL) techniques, and Optimization Algorithms within the realm of these autonomous systems. Drawing from the principles of AI-enhanced methods, the studies encapsulated within this work converge on the exploration and enhancement of different autonomous systems ranging from B5G Truck Platooning Systems, Multi-Agent Systems (MASs), Unmanned Aerial Vehicles, Forest Fire Area Estimation, to the early detection of diseases like Glaucoma. A key research focus, pursued in this work, revolves around the innovative deployment of adaptive PID controllers in vehicle platooning, facilitated through the integration of LLMs. These PID controllers, when infused with AI capabilities, offer new possibilities in terms of efficiency, reliability, and security of platooning systems. We developed a DL model that emulates an adaptive PID controller, thereby showcasing its potential in AI-enabled radio and networks. Simultaneously, our exploration extends to multi-agent systems, proposing an Extended Coevolutionary (EC) Theory that amalgamates elements of coevolutionary dynamics, adaptive learning, and LLM-based strategy recommendations. This allows for a more nuanced and dynamic understanding of the strategic interactions among heterogeneous agents in MASs. Moreover, we delve into the realm of Unmanned Aerial Vehicles (UAVs), proposing a system for video understanding that employs a language-based world-state history of events and objects present in a scene captured by a UAV. The use of LLMs here enables open-ended reasoning such as event forecasting with minimal human intervention. Furthermore, an alternative DL methodology is applied for the estimation of the affected area during forest fires. This approach leverages a novel architecture called TabNet, integrated with Transformers, thus providing accurate and efficient area estimation. In the field of healthcare, our research outlines a successful early detection methodology for glaucoma. Using a three-stage training approach with EfficientNet on retinal images, we achieved high accuracy in detecting early signs of this disease. Across these diverse applications, the core focus remains: the exploration of advanced AI methodologies within autonomous systems. The studies within this thesis seek to demonstrate the power and potential of AI-enhanced techniques in tackling complex problems within these systems. These in-depth investigations, experimental analyses, and developed solutions shed light on the transformative potential of AI methodologies in improving the efficiency, reliability, and security of autonomous systems, ultimately contributing to future research and development in this expansive field. / De Zarzà I Cubero, I. (2023). AI-Enhanced Methods in Autonomous Systems: Large Language Models, DL Techniques, and Optimization Algorithms [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/202201 / Compendio

Abstract The aim of this thesis is to devise new paradigms on radio resource sharing including cache-enabled virtualized large cellular networks for mobile network operators (MNOs). Also, self-organizing resource allocation for small cell networks is considered. In such networks, the MNOs rent radio resources from the infrastructure provider (InP) to support their subscribers. In order to reduce the operational costs, while at the same time to significantly increase the usage of the existing network resources, it leads to a paradigm where the MNOs share their infrastructure, i.e., base stations (BSs), antennas, spectrum and edge cache among themselves. In this regard, we integrate the theoretical insights provided by stochastic geometrical approaches to model the spectrum and infrastructure sharing for large cellular networks. In the first part of the thesis, we study the non-orthogonal multi-MNO spectrum allocation problem for small cell networks with the goal of maximizing the overall network throughput, defined as the expected weighted sum rate of the MNOs. Each MNO is assumed to serve multiple small cell BSs (SBSs). We adopt the many-to-one stable matching game framework to tackle this problem. We also investigate the role of power allocation schemes for SBSs using Q-learning. In the second part, we model and analyze the infrastructure sharing system considering a single buyer MNO and multiple seller MNOs. The MNOs are assumed to operate over their own licensed spectrum bands while sharing BSs. We assume that multiple seller MNOs compete with each other to sell their infrastructure to a potential buyer MNO. The optimal strategy for the seller MNOs in terms of the fraction of infrastructure to be shared and the price of the infrastructure, is obtained by computing the equilibrium of a Cournot-Nash oligopoly game. Finally, we develop a game-theoretic framework to model and analyze a cache-enabled virtualized cellular networks where the network infrastructure, e.g., BSs and cache storage, owned by an InP, is rented and shared among multiple MNOs. We formulate a Stackelberg game model with the InP as the leader and the MNOs as the followers. The InP tries to maximize its profit by optimizing its infrastructure rental fee. The MNO aims to minimize the cost of infrastructure by minimizing the cache intensity under probabilistic delay constraint of the user (UE). Since the MNOs share their rented infrastructure, we apply a cooperative game concept, namely, the Shapley value, to divide the cost among the MNOs. / Tiivistelmä Tämän väitöskirjan tavoitteena on tuottaa uusia paradigmoja radioresurssien jakoon, mukaan lukien virtualisoidut välimuisti-kykenevät suuret matkapuhelinverkot matkapuhelinoperaattoreille. Näiden kaltaisissa verkoissa operaattorit vuokraavat radioresursseja infrastruktuuritoimittajalta (InP, infrastructure provider) asiakkaiden tarpeisiin. Toimintakulujen karsiminen ja samanaikainen olemassa olevien verkkoresurssien hyötykäytön huomattava kasvattaminen johtaa paradigmaan, jossa operaattorit jakavat infrastruktuurinsa keskenään. Tämän vuoksi työssä tutkitaan teoreettisia stokastiseen geometriaan perustuvia malleja spektrin ja infrastruktuurin jakamiseksi suurissa soluverkoissa. Työn ensimmäisessä osassa tutkitaan ei-ortogonaalista monioperaattori-allokaatioongelmaa pienissä soluverkoissa tavoitteena maksimoida verkon yleistä läpisyöttöä, joka määritellään operaattoreiden painotettuna summaläpisyötön odotusarvona. Jokaisen operaattorin oletetaan palvelevan useampaa piensolutukiasemaa (SBS, small cell base station). Työssä käytetään monelta yhdelle -vakaata sovituspeli-viitekehystä SBS:lle käyttäen Q-oppimista. Työn toisessa osassa mallinnetaan ja analysoidaan infrastruktuurin jakamista yhden ostaja-operaattorin ja monen myyjä-operaattorin tapauksessa. Operaattorien oletetaan toimivan omilla lisensoiduilla taajuuksillaan jakaen tukiasemat keskenään. Myyjän optimaalinen strategia infrastruktuurin myytävän osan suuruuden ja hinnan suhteen saavutetaan laskemalla Cournot-Nash -olipologipelin tasapainotila. Lopuksi, työssä kehitetään peli-teoreettinen viitekehys virtualisoitujen välimuistikykenevien soluverkkojen mallintamiseen ja analysointiin, missä InP:n omistama verkkoinfrastruktuuri vuokrataan ja jaetaan monen operaattorin kesken. Työssä muodostetaan Stackelberg-pelimalli, jossa InP toimii johtajana ja operaattorit seuraajina. InP pyrkii maksimoimaan voittonsa optimoimalla infrastruktuurin vuokrahintaa. Operaattori pyrkii minimoimaan infrastruktuurin hinnan minimoimalla välimuistin tiheyttä satunnaisen käyttäjän viive-ehtojen mukaisesti. Koska operaattorit jakavat vuokratun infrastruktuurin, työssä käytetään yhteistyöpeli-ajatusta, nimellisesti, Shapleyn arvoa, jakamaan kustannuksia operaatoreiden kesken.