Tesi per compendi de publicacions. La consulta íntegra de la tesi, inclosos els articles no comunicats públicament per drets d'autor, es pot realitzar prèvia petició a l'Arxiu de la UPC / This doctoral thesis is presented as a compendium of six (6) articles published in Science Citation Index journals. The work proposes harmonic models of some of the most common energy efficient lighting devices in electrical installations (i.e., high intensity discharge lamps (HIDLs) with magnetic ballast, compact fluorescent lamps (CFLs) with integrated ballast, and solid state lighting light emitting diodes (SSL-LEDs) luminaries). The aim of these models and the developed numerical tools is to characterize their harmonic emissions for use in harmonic power flow (HPF) studies in networks with high penetration of these loads. The work contributes mainly with the frequency domain equivalent circuit models and makes significant contributions in the model parameter estimation and suggestions on variations and simplifications of the models. / La presente Tesis Doctoral se defiende bajo la modalidad de un compendio de seis (6) artículos publicados en revistas indexadas en el Science Citation Index. En ella se desarrolla la modelización armónica de algunos de los dispositivos de iluminación de eficiencia energética más habituales en las instalaciones eléctricas (i.e. las lámparas de descarga de alta intensidad con balasto magnético HIDLs, las lámparas fluorescentes compactas con balasto integrado CFLs y las lámparas de estado sólido de diodos inorgánicos emisores de luz SSL-LEDs). El objetivo de estos modelos y de las herramientas numéricas desarrolladas, es caracterizar sus emisiones armónicas y poder utilizarlos en programas de flujos de potencia con armónicos (HPF) en redes eléctricas con gran penetración de este tipo de cargas. El trabajo desarrollado contribuye principalmente con modelos de dichas lámparas basados en su circuito equivalente en el dominio de la frecuencia y realiza aportaciones significativas en la estimación de sus parámetros, además de propuestas de variantes y simplificaciones de los modelos.

Current transport network owners are focused on offering services on top of the infrastructures they own, while end users have no control over them. Traditionally, this has been their business model, as the cost of building the infrastructures to provide services is considerably high. However, the traffic on Internet has been, and still is, rapidly increasing over the years. Additionally new emerging services are pushing the limits of existing telecommunication infrastructures, particularly transport optical networks. To overcome such situation, network virtualization has been considered as an effective solution for the future optical networks architectures. Thanks to Virtual Optical Networks (VONs), it is possible to create mission-specific logic infrastructures, which fulfil the exact requirements of the applications that will run on top of them, sharing a unique physical substrate. However, the applicability of virtualization techniques to the optical domain is still under research, being on key point the mapping of the virtual resources to the actual physical ones. However, virtualization per se does not provide a solution flexible enough in terms of bandwidth utilization. For this to happen, an equally flexible transport technology must be adopted. Elastic Optical Networks (EONs) have been presented as an efficient solution for flexible bandwidth allocation. Additionally, due to the dinamicity of the traffic patterns that such virtual networks will face, it is highly desirable to provide a physical substrate that will help on keeping the associated operational expenditures (OPEX) at low levels, being a very important parameter the energy consumption. The energy consumption topic has been subject of big research efforts in order to provide more energy efficient optical transport networks, which, at their turn, will help on the creation of less costly virtual infrastructures. This thesis is devoted to the study of resource allocation to VONs, aiming to provide a flexible, efficient and optimized environment for the embedding of the VONs to the actual physical substrate. The considered scenario is composed of an underlying optical transport network and multiple client VONs that have to be allocated on top. In such scenario, a key aspect relates to how actual resources are associated to the virtual ones, guaranteeing the isolation among VONs and satisfying the resources requirements of every one of them. After an introduction to the thesis, chapter 2 surveys nowadays optical network infrastructures, concluding on the need to move towards a more dynamic and efficient optical network infrastructure. Next, it proceeds to summarize the state of the art of the concepts that enable for such network architecture, namely, VONs, EONs and energy efficient optical infrastructures. Then, chapters 3, 4 and 5 focus on providing solutions to optimize specific aspects of these enabling concepts. More in details, chapter 3 studies the main challenges on the VON embedding problem and presents solutions that allow for an optimized resoure assignment to VONs in a physical substrate depending on the VONs characteristics and the sppecific network substrate. Chapter 4 proposes the Split Spectrum (SS) approach as a way to improve the spectrum utilization of EONs. Finally, chapter 5 focuses on provide and evaluate routing and architectural solutions in aims to reduce the energy consumption of the optical substrate so as VONs with lower OPEX can be deployed on top of it. / Els actuals propietaris de xarxes de transport es centren en oferir serveis mitjançant les infraestructures que posseeixen, mentre els usuaris finals no tenen cap control sobre aquests. Tradicionalment, aquest ha estat el seu model de negoci, ja que el cost de construir aquestes infraestructures és considerablement elevat. Tanmateix, el tràfic a Internet ha estat creixent de manera ràpida durant els últims anys. A més, l'aparició de nous serveis està portant al límit les actuals infraestructures de telecomunicacions, especialment les xarxes òptiques de transport. Per tal de superar aquesta situació, la virtualització de xarxes és considerada com una solució efectiva per les futures arquitectures de xarxes òptiques. Gràcies a les Xarxes Òptiques Virtuals (VONs), és possible crear infraestructures lògiques específiques en la seva missió, les quals permeten satisfer els requisits de les aplicacions que s'executaran a través d'elles, compartint un únic substrat físic. Tanmateix, l'aplicació de les tècniques de virtualització en el domini òptic encara és subjecte d'investigació, sent el mapeig entre els recursos virtuals i els recursos físics un punt clau que cal adreçar. No obstant això, la virtualització en si mateixa no proporciona una solució prou flexible en termes d'utilització de l'espectre. Per aquest motiu és necessari que el substrat físic adopti una tecnologia igualment flexible. Les Xarxes Òptiques Elàstiques (EONs) es presenten com una solució eficient per a una assignació flexible de l'espectre. A més, a causa del dinamisme dels perfils de trafic als quals s'enfrontaran les VONs, és desitjable proporcionar una infraestructura física que ajudi a mantenir baixes les despeses operatives (OPEX) d'aquestes xarxes, sent un paràmetre molt important el consum energètic. El tema del consum energètic ha estat subjecte de grans iniciatives de recerca per tal de proporcionar xarxes de transport òptiques més eficients energèticament, les quals permetran crear VONs menys costoses. Aquesta tesi està dedicada a l'estudi l'assignació de recursos a les VONs, amb l'objectiu de proporcionar un entorn flexible, eficient i optimitzat per a la incrustació de les VONs al substrat físic. L'escenari considerat es compon d'una xarxa de transport subjacent i múltiples VONs client a col·locar sobre el substrat físic. En aquest escenari, un aspecte clau es refereix a com els recursos reals s'associen als virtuals, garantint l'aïllament entre VONs i satisfent els recursos demanats per cada una d'elles. Després d'una introducció a la tesi, el capítol 2 revisa les infraestructures de xarxa òptica actuals, concloent en la necessitat d'avançar cap a infraestructures més dinàmiques i eficients. Tot seguit, es procedeix a resumir l'estat de l'art dels conceptes que habilitaran aquesta arquitectura de xarxa, bàsicament, VONs, EONs i les xarxes òptiques de baix consum. A continuació, els capítols 3, 4 i 5 es centren en proporcionar solucions per optimitzar aspectes específics d'aquests conceptes. Més en detall, el capítol 3 estudia els principals reptes en el problema de la incrustació de VONs i presenta solucions que permetin assignar recursos de manera optimitzada a les VONs en un substrat físic. El capítol 4 proposa el concepte de l'Split Spectrum (SS) com una forma de millorar la utilització de l'espectre en les EONs. Finalment, el capítol 5 es centra en proporcionar i avaluar solucions arquitectòniques i d'enrutament amb l'objectiu de reduir el consum d'energia del substrat òptic de tal manera que VONs amb menor OPEX puguin ser desplegades a través d'ell. / Los actuales propietarios de las redes de transporte se centran en ofrecer servicios mediante las infraestructuras que poseen y gestionan, mientras que los usuarios finales no tienen ningún control sobre estos. Tradicionalmente, este ha sido el modelo de negocio adoptado por los operadores de redes, ya que el coste de construir y mantener las infraestructuras correspondientes por tal de ofrecer servicios mediante ellas era, y aun es, considerablemente elevado. No obstante, el tráfico en Internet ha crecido de manera rápida y sostenida durante los últimos años y se prevé que continuara con este crecimiento en el futuro. Además, la aparición de nuevos servicios y paradigmas, están llevando al límite las actuales infraestructuras de telecomunicaciones, especialmente las redes de trasporte óptico. Por tal de superar dicha situación, la virtualización de redes ha sido considerada como una solución efectiva para las futuras arquitecturas de redes ópticas. Gracias a las Redes Ópticas Virtuales (VONs), es posible crear infraestructuras lógicas especificas en su misión, las cuales podrán satisfacer los requisitos de las aplicaciones que se ejecutaran a través de ellas, usando y compartiendo un único sustrato físico. No obstante, la aplicación de las técnicas de virtualización en el dominio óptico aun es sujeto de investigación, siendo el mapeo entre los recursos virtuales y los físicos (también conocido como incrustación de la red virtual) un punto clave a solucionar. No obstante, la virtualización por si misma no ofrece una solución suficientemente flexible en términos de utilización del ancho de banda. Por tal de proporcionar un entorno de virtualización suficientemente flexible para acomodar cualquier ancho de banda con suficiente granularidad, es necesario que el sustrato físico adopte una tecnología de transporte igual de flexible. Las Redes Ópticas Elásticas (EONs) se presentan como una solución eficiente para una asignación flexible del ancho de banda en redes ópticas. Además, debido a la heterogeneidad y dinamismo de los perfiles de tráfico a los cuales se enfrentaran las redes virtuales, es altamente deseable proporcionar una infraestructura física que ayuda a mantener bajos los gastos operativos (OPEX) de estas redes, siendo un parámetro muy importante el consumo energético asociado a la operación de las VONs. El tema del consumo energético ha sido, y aun es, sujeto de grandes iniciativas de investigación centradas en desarrollar nuevas arquitecturas de dispositivos o algoritmos de asignación de recursos conscientes del consumo energético por tal de proporcionar redes de transporte ópticas más eficientes energéticamente que, a su vez, permitan crear infraestructuras virtuales menos costosas des del punto de vista energético. Esta tesis se centra en el estudio de la composición y asignación de recursos a las VONs, con el objetivo de proporcionar un entorno flexible, eficiente y optimizado para la incrustación de las VONs en el sustrato físico real. El escenario considerado se compone de una red de transporte subyacente, ya sea una Red Óptica de Conmutación de Longitud de Onda (WSON) o EON, y múltiples VONs cliente, las cuales se colocaran encima del sustrato físico. En este escenario, un aspecto clave se refiere a como los recursos reales se asocian a los virtuales, garantizando el aislamiento entre VONs y satisfaciendo los recursos pedidos (por ejemplo, capacidad de enlace) por cada una de ellas. Después de una introducción a la tesis, el capítulo 2 revisa las infraestructuras de redes ópticas actuales, concluyendo en la necesidad de avanzar hacia una infraestructura de red óptica más dinámica y eficiente por tal de afrontar el crecimiento del tráfico en Internet y la aparición de nuevos servicios y paradigmas. Seguidamente, se procede a resumir el estado del arte de los conceptos y paradigmas que permitirán habilitar esta arquitectura de red, básicamente, VONs, EONs y las infraestructuras ópticas de bajo consumo energético. A continuación, los capítulos 3, 4 y 5 se centran en proporcionar soluciones para optimizar aspectos específicos de estos conceptos con la finalidad de proporcionar un marco optimizado que ayudara en la configuración de las futuras infraestructuras de redes ópticas y sus modelos de negocio. Concretamente, el capítulo 3 estudia los principales retos en el problema de la incrustación de VONs y presenta soluciones que permiten una asignación de recursos optimizada a las VONs en un sustrato físico dependiendo de las características de las VONs y del sustrato de red. El capítulo 4 propone el concepto de Split Spectrum (SS) como una forma de mejorar la utilizaci_on del espectro en las EONs. Finalmente, el capítulo 5 se centra en proporcionar y evaluar soluciones arquitectónicas y de enrutamiento con el objetivo de reducir el consumo energético del sustrato óptico de tal manera que VONs con menor OPEX puedan ser desplegadas mediante este sustrato.

Class-D Amplification has been an absolute improvement in terms of electrical efficiency in audio amplification systems, but it has been during the first stabilization decade of this technology in the market when manufacturers have made public their concern about the incompatibility between distributed audio or PA (Public Address) applications and HiFi or Professional sound applications. High quality sound systems need switching frequencies (>1MHz) whose switching losses do not allow to work at high power (>250W). To the contrary, applications for high power amplification need high electrical efficiency, so the switching frequencies must be set to low values (<400kHz), reducing the signal quality. This incompatibility is called Power-Quality Compromise. The Characteristics of the Multilevel Topologies of DC-AC power electronic converters allow them to work at high power (increasing the working voltage) while maintaining or even increasing the quality of the output signal using the extra levels in the PWM output signal. The application of these topologies in the world of sound amplification could lead to the resolution of the Quality-Power Compromise and so, the main problem of the Class-D amplifiers. In the first place, this Thesis proposes the study of the Multilevel Topologies in order to determine the one who shows better features and feasibility, in terms of implementation as a Switching Audio Amplifier. Having determined the Optimal Topology, it is proposed to study, characterize and model the non-ideal phenomena and those mechanisms who generate Time and Pulse Width errors in the PWM output signals of the Amplifier Topology selected. Using these models, the effects of the errors in the output signals are studied by means of computer simulation, expressing these effects using non normative merit factors. These merit factors, unlike regulations, allow to evaluate quantitatively and qualitatively the degree of involvement of each one of the different error sources intrinsic to the design and topology. From these results it can be developed Estimation Models of the Non Normative Merit Factors, which allow to calculate the values of the Width Error and Distortion, local or total, generated by any source of error in the Amplification Sound Chain. In order to express this information in normative values, Estimation Models of Electric Efficiency, Distortion (THD) and Signal Efficiency are developed from the estimated non normative values of Width Error Distortion. These estimation models provide an information that, once ordered and classified in order to be used as a tool in the different levels of the design process, production and marketing of Switching Amplifier Equipment, allow the optimal component selection in the pre-design phase and the decision-making process to ensure the future validation of the Multilevel Switching Amplifier under market regulatory criteria. The Estimation models of the Width and Distortion Errors and the estimation models of the Electrical Efficiency, Distortion and Signal have been validated through the comparison with experimental results from a laboratory prototype and an industrial manufacturing prototype, designed by a company in the sector using the Optimal Design Tools based on the Estimation Models developed in this Thesis. The experimental results confirm that the normative values measured of the Electrical Efficiency, Distortion and Signal Efficiency are within the acceptable values for the commercialization of the equipment. / L'amplificació en Classe-D ha suposat una millora absoluta en quant a eficiència elèctrica en els sistemes d'amplificació d'àudio, però ha estat durant la primera dècada d'estabilització d'aquesta tecnologia al mercat, quan els fabricants han fet pública la seva preocupació per la incompatibilitat entre aplicacions de so distribuït o PA (Public Addres) i so HiFi o Professional. Els sistemes d'alta qualitat de so, necessiten de freqüències de commutació (>1MHz) per les quals les pèrdues en commutació no permeten treballar a potències elevades (>250W). Per al contrari, per aplicacions d'amplificació de so de potència es necessita d'alta eficiència elèctrica, amb el que les freqüències de commutació han de quedar fixades a valors baixos (<400kHz), reduint la qualitat del senyal. Aquesta incompatibilitat d'aplicacions s'anomena Compromís Qualitat-Potència. Les topologies Multinivell de convertidors electrònics de potència DC-AC, per les seves característiques, permeten treballar a gran potència (augmentant la tensió de treball), mantenint o augmentant la qualitat del senyal de sortida mitjançant els nivells extra en el senyal PWM de sortida. L'aplicació d'aquestes topologies en el mon de l'Amplificació de So, podria suposar la resolució del compromís Qualitat-Potència, i per tant de la principal problemàtica que presenta la Classe-D. Aquesta Tesis proposa en primer lloc l'estudi de les topologies Multinivell per determinar quina és la que millors prestacions i viabilitat mostra en quant a la seva aplicació com a Amplificador d'Àudio Commutat. Un cop determinada la Topologia Òptima, es proposa estudiar, caracteritzar i modelar els fenòmens no ideals i mecanismes generadors d'errors de Temps i d'Amplitud al polsos del senyal PWM de sortida d'un Amplificador amb la Topologia seleccionada. A partir d'aquests models, s'estudia mitjançant simulació els efectes dels errors en els senyals de sortida, expressant aquests efectes mitjançant factors de mèrit no normatius. Aquests factors de mèrit, al contrari que els normatius, permeten avaluar quantitativa i qualitativament el grau d'afectació de cada una de les diferents fonts d'errors intrínseques al disseny i a la topologia. A partir d'aquests resultats es desenvolupen Models d'Estimació dels Factors de Mèrit No Normatius, que permeten calcular el valor d'Error d'Amplitud i Distorsió, local o total, generat per qualsevol font d'error de la Cadena d'Amplificació de So. Per tal d'expressar aquesta informació en valors normatius, es desenvolupen Models d'Estimació d'Eficiències Elèctriques, Distorsió (THD) i Eficiència de Senyal a partir dels valors estimats d'Error d'Amplitud i Distorsió no normatius. Aquests models d'estimació proveeixen d'una informació que, un cop ordenada i classificada per tal de ser utilitzada, com a eina en els diferents nivells del procés de disseny, producció i comercialització d'equips d'amplificació commutada, permet, de forma òptima en la fase de pre-disseny, la selecció de components i la presa de decisions per tal d'assegurar, un cop construït, la futura validació de l'amplificador commutat multinivell, sota criteris normatius de mercat. Els models d'estimació d'Errors d'Amplitud i Distorsió, així com els models d'estimació de les Eficiències Elèctriques, de Distorsió i de Senyal han estat validats mitjançant la seva comparació amb resultats experimentals obtinguts a partir d'un prototipus de laboratori i un prototipus de fabricació industrial, dissenyat per una empresa del sector a partir de l'ús de les Eines de Disseny Òptim basades en els Models d'Estimació, desenvolupades en aquesta Tesis. Els resultats experimentals confirmen que els valors normatius mesurats d'Eficiència Elèctrica, Distorsió i Eficiència de Senyal estan dintre dels valors acceptables per la comercialització de l'equip.

It is widely believed that many drawbacks in today's wireless communication paradigm might be relieved by enabling high carrier frequency transmission, and endowing both the network and the user equipment with some degree of reconfigurability. The urgency of a new framework in wireless digital transmission which should allow for higher bit rate, lower latency and tighter delay constraints, led us to investigate the fundamental building blocks which, at the circuital/device level, will foster a change towards more efficient communication schemes, delivering a more satisfactory end user experience. Specifically, this work deals with the inherently analog devices, found at the core of each transceiver module and capable of providing the carrier signal; these are the oscillators. In particular, two distinct classes of oscillators are regarded central to our contribution. One class is constituted by N-push oscillators, which thanks to coupling effect of N identical core oscillators allow N-fold harmonic generation (and thus high frequency transmission). The second class is constituted by wideband tunable oscillators, whose topology derives from a feedback distributed amplifier and therefore called distributed oscillators; by adequately altering the bias level at each section Distributed Voltage Controlled Oscillators can be implemented (which can scan a wide frequency range). The introductory part of this work, deals with their operation principles in great detail. As microwave oscillators are nonlinear devices, a full nonlinear analysis, synthesis, and optimization is considered for their implementation. Consequently, nonlinear numerical techniques have been reviewed in the second part of the thesis. Particularly, the role of Harmonic Balance simulations and the auxiliary generator/probe method for obtaining the oscillator solutions has been emphasized; the overall research goal of this dissertation is to show that the former techniques are very effective in obtaining detailed information about the periodic steady state behavior for the two class of circuits being investigated. A triple-push oscillator topology has been initially considered. Provided a certain phase distribution is maintained among the oscillating elements, the output power of the third harmonic increases while the lower order harmonics cancel out, which represents the default operating mode. Due to circuit symmetry, to the presence of delay in the coupling network and to unavoidable mismatches, unwanted oscillating modes might coexist with the intended one. A design strategy relying on the Harmonic Balance parametric analysis of the oscillating voltage at a selected node in the coupling network with respect to coupling phase and coupling strength is presented, to the aim of quenching undesired oscillation modes. Moreover the design of a four stage reverse mode distributed voltage controlled oscillator (DVCO) has been described. All the design steps have been reported, from a very idealized, purely behavioral design to a very concrete one, involving details derived from electromagnetic simulations. Harmonic Balance techniques were used to evaluate its tuning function, output power and DC current consumption, which have been completely characterized across the tuning bandwidth. Finally, a method for an optimized design with reduced variations in the output power has been presented. An alternative implementation, targeting wider tuning ranges/ higher oscillation frequencies was introduced. The measurements performed on the fabricated prototypes revealed good agreement with the simulation results, confirming the validity of the approach.

This Thesis proposes the design and the optimization of a hybrid WDM/TDM PON at the L1 (PHY) and L2 (MAC) layers, in terms of minimum deployment cost and enhanced performance for Greenfield NGPON. The particular case of RSOA-based ONUs and ODN using a single-fibre/single-wavelength is deeply analysed. In this WDM/TDM PON relevant parameters are optimized. Special attention has been given at the main noise impairment in this type of networks: the Rayleigh Backscattering effect, which cannot be prevented. To understand its behaviour and mitigate its effects, a novel mathematical model for the Rayleigh Backscattering in burst mode transmission is presented for the first time, and it has been used to optimize the WDM/TDM RSOA based PON. Also, a cost-effective, simple design SCM WDM/TDM PON with rSOA-based ONU, was optimized and implemented. This prototype was successfully tested showing high performance, robustness, versatility and reliability. So, the system is able to give coverage up to 1280 users at 2.5 Gb/s / 1.25 Gb/s downstream/upstream, over 20 Km, and being compatible with the GPON ITU-T recommendation. This precedent has enabled the SARDANA network to extend the design, architecture and capabilities of a WDM/TDM PON for a long reach metro-access network (100 km). A proposal for an agile Transmission Convergence sub-layer is presented as another relevant contribution of this work. It is based on the optimization of the standards GPON and XG-PON (for compatibility), but applied to a long reach metro-access TDM/WDM PON rSOA-based network with higher client count. Finally, a proposal of physical implementation for the SARDANA layer 2 and possible configurations for SARDANA internetworking, with the metro network and core transport network, are presented.

Electro-optical traveling wave modulators (EO-TWM) are basic building blocks of the optical communications industry which is leading a revolution in the way we communicate, work and live. As a result, the demand for high-speed data transmission with low driving voltage is continuously growing up with costs that should be kept below a minimum. Besides communications, a growing number of applications for EO-TWM is continuously emerging with equally stringent requirements. This Thesis is concerned with advances in the eld of systematic design and optimization of EO-TWM for coping inverse of the velocity matching constant has been shown to govern the low-loss limit (LL), while in the velocity matching limit (VM), a constant bandwidth times squared-length rule proportional to the inverse of the squared loss constant has been found more appropriate. In this work we provide insights into the trade-o issue in EO-TWM, and a complete picture of the applicable gures of merit for every operative range. Besides the known LL and VM gures of merit, two intermediate ranges, the quasi-low loss (QLL) and the quasi-velocity matching (QVM), have been identi ed. Also novel closed-forms expressions fully accounting for the e ffects of the skin-e ffect electrode loss and optical-electrical wave velocity mismatch, explicitly relating the operative bandwidth and the electrode length in EO-TWM, have been found. Novel bandwidth and electrode-length charts have been created, which constitute a useful tool for the optimization and design of this modulators. A graphical interface tool called MZM-GIT has been built integrating the analytical optimization and design strategies developed throughout the Thesis. With the aid of the MZM-GIT, several proposals of optimized MZM designs based on practical structures described in literature, and also based on the industry trends, are made and analyzed. with the industrial demands. In EO-TWM, the accumulated electro-optic e ect over the optical wave grows with the co-propagated traveling wave (TW) length, allowing to reduce the required RF driving power. However, in typical electro-optic materials for modulators, among which LiNbO3 stands up, due to the natural mismatch between the velocity of the RF and the optical waves, the modulation bandwidth decreases with the TW length, giving place to a well-known trade-o ff. In typical LiNbO3 substrates, in which this Thesis is focused, this trade-off is seen to mainly depend on the values of the electrical loss constant and the e ective wave velocity mismatch in the TW structure that forms the electrodes, usually a coplanar waveguide (CPW). Special emphasis has historically been placed on the optimized design of the CPW in EO-TWM. In this Thesis the study of closed-form expressions for the propagation parameters of CPW as a function of the geometry, has proven useful for the design and optimization procedures sought. Although some interesting approaches to closed-form formulations have been found in literature, none of them completely ful lls the desired requirements of providing a reliable yet simple description of propagation in CPW, appropriate to systematic and easy to follow design rules for EO-TWM, and therefore new simpli ed closed-form expressions for the CPW transmission parameters have been developed. In a second part of the Thesis, the bandwidth-length trade-off has been examined. To date, two bandwidth-length rules have been proposed: a constant bandwidth-length product proportional to the inverse of the velocity matching constant has been shown to govern the low-loss limit (LL), while in the velocity matching limit (VM), a constant bandwidth times squared-length rule proportional to the inverse of the squared loss constant has been found more appropriate. In this work we provide insights into the trade-off issue in EO-TWM, and a complete picture of the applicable fi gures of merit for every operative range. Besides the known LL and VM gures of merit, two intermediate ranges, the quasi-low loss (QLL) and the quasi-velocity matching (QVM), have been identi ed. Also novel closed-forms expressions fully accounting for the e ffects of the skin-eff ect electrode loss and optical-electrical wave velocity mismatch, explicitly relating the operative bandwidth and the electrode length in EO-TWM, have been found. Novel bandwidth and electrode-length charts have been created, which constitute a useful tool for the optimization and design of this modulators. A graphical interface tool called MZM-GIT has been built integrating the analytical optimization and design strategies developed throughout the Thesis. With the aid of the MZM-GIT, several proposals of optimized MZM designs based on practical structures described in literature, and also based on the industry trends, are made and analyzed.

This thesis is framed in the field of mobile communications and more particularly in handset antennas. The wireless industry is constantly growing, which entails challenging handset antenna specifications. Handset antennas not only have to be multi-band for satisfying the great number of communication services, but also sufficiently small as for fitting in the reduced space imposed by the handset platforms. The appearance of the MIMO (Multiple Input Multiple Output) technology, further exacerbates these challenges. In order to satisfy these requirements, this thesis proposes the use of the ground plane, inherently present in any handset platform, as the main radiator. Electrically small nonresonant elements, called along this thesis as ground plane boosters, are used to transfer energy to this ground plane. The solution removes the need of including a dedicated antenna featured by considerable dimensions, thus releasing space to integrate other antennas, as well as, other handset components, services and functionalities.

Terahertz is the portion of the spectrum that covers a frequency range between 300 GHz - 3 THz. This frequency band has proven its potential for imaging applications thanks to the good compromise between spatial resolution and penetration; however, this push towards high frequencies contains many technological difficulties in all the subsystems involved in the signal generation, transmission and detection. The power budget restrictions and high losses that sources and receivers currently suffer at these frequencies require systems with a high level of integration among all the devices and components of the systems and subsystems. Therefore, the antennas needed for these systems require to be integrated within the same fabrication processes and technologies as the sensing and power converting devices that are used at their terminals. This doctoral thesis has focused on the development of integrated antenna arrays at Terahertz frequencies for imaging applications, for both near-field and focal-plane geometries, with a special emphasis on the technologies and the fabrication capabilities that can be potentially used and are currently available. The current imaging systems require large arrays of antennas in order to achieve the high-speed image acquisition that is required in most THz applications. This fact increases considerably the difficulty and complexity to achieve highly integrated and efficient antennas. This thesis has characterized and analyzed these difficulties and provided solutions to the development of antenna arrays at millimeter and submillimeter wave frequencies. The first part of this thesis has focused on the study of a planar antenna array, called retina, for a specific near-field imaging system based on the Modulated Scatterer Technique (MST) at millimeter and submillimeter-wave frequencies. This system has been selected for its capabilities to perform high-speed imaging and because it does not require a high frequency distribution line network. However, it is hindered by many technological difficulties: the selection of an antenna geometry that achieves high efficiency, the selection of the adequate active element and its integration with the antenna. In this thesis, these challenges have been addressed and studied in-depth, and a design methodology that integrates all the different aspects of the system has been developed. Because planar antennas at millimeter and submillimeter wave frequencies suffer from high losses due to the surface wave modes inside substrate, these losses have been analyzed and quantified for different antennas, and an antenna geometry that reduces significantly this problem has been developed. Different switching technologies currently or potentially available at these high frequencies have been considered in order to study and analyze their capabilities and their integration possibilities: PIN diodes, Schottky diodes and RF-MEMS. These technologies have been studied through the development of three retina prototypes, which have been fabricated using high precision fabrication processes such as laser micromachining and photolithographic. Different measurement set ups were fabricated and assembled to validate the different premises presented. The second part of the thesis is devoted to the study of integrated Focal Plane Arrays (FPA). The development of FPA at submillimeter wave frequencies is highly on demand due to the enormous progress in designing integrated heterodyne receivers. These receivers integrate arrays of submillimeter-wave diode-based mixers and multipliers with Monolithically Integrated Circuit (MIC) amplifiers on the same wafer stack. For this stackable multi-pixel terahertz camera technology to work, a leaky wave antenna with silicon micro-lenses has been developed, which allows wafer level integration compatible with silicon micro-fabrication techniques for bulk array manufacturing and has high directivity in order to illuminate a reflector efficiently. Detailed and thorough design guidelines for this antenna are presented. Two antenna prototypes were built in order to evaluate the two fabrication possibilities: advanced laser micro-fabrication and photolithographic fabrication. A study of the aberrations of the lens has been developed in order to evaluate the performance of the lens profile fabricated. Moreover, a set of radiation pattern measurements of the fabricated prototypes was performed in order to evaluate the performance of the antenna and its possibilities to be used as a FPA.

El objetivo último de este trabajo es el estudio de la degradación de las células solares en el ambiente espacial fuera de la atmósfera, con el fin de conseguir mejores rendimientos durante el mayor tiempo posible de los generadores fotoeléctricos en vehículos espaciales, y tener la capacidad de hacer predicciones a largo plazo del comportamiento de las células.

This research is developed in the context of the Next Generation of Passive Optical Networks (NG-PON) for Fibre to the Home (FTTH) network. In this thesis, several techniques are proposed in order to increase the network capacity, dealing with the main optical impairments present in PON with minimum infrastructure, and with the goals of wavelength transparency and cost effectiveness. The main issues in this work are the photodetection process over chromatic dispersion (CD) and ASE noise impairments, the mitigation strategies for Rayleigh Backscattering (RB) effect and the WDM-PON RSOA-based ONUs, with their limitations of bandwidth and chirp. Regarding photodetection, a novel module which performs the mathematical function SQRT has demonstrated its linearizing capability at partially compensating the square-law of the photon to electron conversion process after reception. Improvements in RoF systems in terms of harmonics level reduction and in digital transmissions with linear FFE equalizers have been achieved. Also, the complexity requirements in MLSE receivers are reduced after the SQRT module. This module has been implemented electronically, first, by using the intrinsic non-linearity of the schottky diodes, on a microstrip circuit. After that, a MMIC chip has been designed and experimentally tested. But the SQRT module is also useful for other applications as downstream cancellation in the upstream signal in bidirectional transmissions, and to reduce the amplitude differences of the different packets in burst mode receivers, relaxing the adjustment requirements for the following limiting amplifiers. The statistics of the received signals in optically amplified (ASE dominant) systems have been also studied. The optimum function, capable of recovering the Gaussian distribution after detection, has been obtained. For systems working at FEC limit, like with MLSE, the improvement over SQRT is small, validating the good performance of the SQRT module. RB effect has been mitigated by using wavelength shifting at the ONU premises. Several wavelength shifter devices have been proposed for the first time, based in 2 and 3 arms capable of translating (in several GHz) the wavelength that comes from the OLT to avoid the overlap between the upstream signal and the RB, while maintaining the signal in the same AWG channel. These devices are less complex than the recently proposed 4-arm structures and require simpler signals than the sawtooth that is necessary for a 1-arm phase modulator. Experimental validation of the WS technique for RB mitigation has been performed, first with a commercial dual arm MZM and after by using a RSOA with a low duty square signal with improvements of 5 to 7dB in tolerance to RB. The new standards that have just been approved (XG-PON, 10G-EPON) propose the upgrading of the network towards symmetrical 10G/10G operation. However, RSOA’s bandwidth and chirp still limit the performance at these bit rates. The first 10Gb/s transmission in real time and with low BER (10-9) using a RSOA as an intensity modulator has been reported by using some of the techniques presented in this thesis: duobinary transmission and detuned optical filtering plus DFE-FFE equalization. Other techniques proposed and experimentally tested are the phase adjustment with a programmable optical filter and the direct detection together with MLSE equalization. / Aquest treball de recerca està desenvolupat en el context de la nova generació de xarxes òptiques passives (NG-PON) de fibra fins a la llar (Fibre to the Home (FTTH)). En aquesta tesi, s’han proposat diverses tècniques que permeten augmentar la capacitat de la xarxa, tractant les principals degradacions presents en les xarxes PON amb la mínima infrestuctura, amb els objectius de mínim cost i transparència front a la longitud d’ona triada per a la transmissió. Els principals temes d’aquest treball són el procés de fotodetecció després d’una transmissió amb dispersió cromàtica (CD) i soroll d’emissió espontània (ASE), les estratègies per combatre l’efecte de Rayleigh Backscattering (RB) en la trasmissió bidireccional amb la mateixa longitud d’ona i les xarxes PON basades en unitats d’usuari (ONU) que utilitzen l’ amplificador semiconductor reflexiu (RSOA) com a transmissor, amb les seves limitacions d’ample de banda i chirp. Pel que fa a la fotodetecció, un dispositiu novedós que realitza la funció matemàtica arrel quadrada (SQRT) ha demostrat la seva capacitat de linearització al compensar parcialment la funció quadrada del procés de conversió de fotó a electró després de recepció. S’ha aconsegit millorar les prestacions en xarxes de ràdio sobre fibra en termes de reducció del nivel d’harmònics i també en transmissió de senyals digitals amb equalitzadors electrònics lineals. La complexitat de l’ equalitzador de màxima verosimilitud, o MLSE, es redueix si s’utilitza el mòdul SQRT. Aquest dispositiu s’ha implementat electrònicament en un circuit microstrip amb dos diodes schottky i també en un xip MMIC . El bloc SQRT també és útil per aplicacions de cancel•lació del senyal de baixada present en el senyal de pujada en les transmissions bidireccionals i per igualar l’amplitud dels paquets en senyals que es reben a ràfegues (burst mode). L’estadística del senyal rebut en sistemes amplificats òpticament (on domina el soroll ASE) també ha estat estudiat. La funció òptima, capaç de recuperar la distribució estadistica Gaussiana després de detecció, ha estat obtinguda. Per sistemes que treballen amb un nivell de errors en el límit FEC, com el MLSE, la millora de la funció òptima és molt petita front la SQRT, demostrant el bon funcionament del mòdul SQRT. L’efecte de RB s’ha minimitzat utilitzant la tècnica de desplaçar la longitud d’ona del senyal de baixada a la ONU. S’han disenyat diversos dispositius electro-òptics capaços de realitzar aquest desplaçament, utilitzant unes configuracions novedoses basades en 2 i 3 branques. L’efecte RB és redueix evitant el sol•lapament espectral però mantenint la longitud d’ona dins el mateix canal del AWG. Aquests dispositius són més senzills que les recentment proposades estructures basades en 4 branques i requereixen senyals més simples que el dent de serra necessari per al modulador de fase d’una branca. La tècnica WS s’ha validat experimentalment usant un MZM comercial i també usant un RSOA, aconseguint-se unes millores de 5 a 7 dB en tolerància a RB. Els nous estàndards que han estat recentment aprovats (XG-PON, 10G-EPON) proposen l’evolució de la xarxa passiva cap a la transmissió simètrica a 10Gb/s. La realitat és, però, que l’ample de banda i el chirp limiten la utilització de RSOAs a aquesta velocitat. La primera transmissió a 10Gb/s en temps real amb baixa tasa d’errors (BER=10-9) usant el RSOA com a modulador d’intensitat a la ONU ha estat presentat utilitzant les tècniques explicades en aquesta tesi: modulació duobinaria i filtrat òptic desplaçat amb equalització electrònica (DFE-FFE). Altres mètodes utilitzats han estat l’adjust de la fase amb un filtre òptic programable i el MLSE.