M.Ing. / This work considers a dual band electronically steerable parasitic array radiator (ESPAR) antenna system. This system is designed with one actively fed radiating element and N-parasitic radiating elements. The radiation pattern can be electronically controlled by means of the loads terminating the parasitic elements. The antenna system is designed as ESPAR to have a minimum number of controlling elements thereby minimising the power consumed. The dual band operation of this antenna is for the frequency bands of the wireless local area network (WLAN), which are 2.412-2.482 GHz for IEEE 802.11b/g (known as the 2.4 GHz band) and 5.15-5.825 GHz for IEEE 802.11a (known as the 5 GHz band). In the upper band, only the 5.8 GHz sub-band was considered. The dual-band capability was targeted by conducting a structural parameter modification on the antenna system. The structural modification involves optimisation of the length of the active element, the length of the parasitic element, the distance of the parasitic element from the active element and most importantly, by application of a loading technique on the elements. The loading was done by using optimisation tools, such as fminsearch, fminbnd and the genetic algorithm. The specific circuit that was used for the loading was a series connection inductors inserted into the antenna’s elements at positions found via a global optimization.. The method used was to first identify the optimal length per specific resonant frequency and consider the optimal length with respect to both resonant frequencies. The second step was to load the three resulting optimised different monopoles, and the loading with results closest to the requirements. The optimum monopole of the three in the second step was then used as the fixed input parameter for the main optimisation of ESPAR antenna. Using a ground plane with a skirt, an acceptable return loss performance has been achieved for the antenna's main building M.J. Bembe ii block, a monopole element, in both frequency bands. The challenge was found in steering of the beam in different directions; it was then concluded that the usage of more elements could provide the necessary freedom for the optimisation process. Six elements were arranged symmetrically close to the active fed element in order to achieve a dual band resonance, with different designs meeting the requirements differently. This is the first report showing an ESPAR antenna optimisation which includes the loading of elements with lengths and distance optimisation.

Les systèmes de communications sans fil n'ont cessé d'évoluer ces dernières années, poussés par de fortes demandes du marché en systèmes toujours plus autonomes et performants. Ainsi, de nouvelles contraintes de conception sont apparues de manière à mieux prendre en compte les aspects énergétiques et ainsi améliorer la durée de vie des batteries et des circuits. Actuellement, les systèmes de communications numériques sans fil consomment d'importantes quantités d'énergie. D'autre part, la complexité des systèmes croît de génération en génération afin de satisfaire toujours plus d'utilisateurs avec un haut niveau de performances. Dans ce contexte à fortes contraintes, les circuits de type FPGA apparaissent comme une technologie attractive, pouvant supporter des circuits numériques complexes grâce à leur grand nombre de ressources. Pour pouvoir concevoir les futurs systèmes de communications numériques sans fil sur ce type de circuit, les concepteurs de tels systèmes doivent pouvoir estimer la consommation et les performances au plus tôt dans la phase de conception. De cette façon, ils pourront explorer l'espace de conception et effectuer des choix d'implémentation afin d'optimiser leurs systèmes. Durant cette thèse, une méthodologie a été proposée dont les objectifs sont d'estimer rapidement et à haut niveau la consommation de leurs circuits implantés sur FPGA ainsi que leurs performances, d'explorer l'espace de conception, de comparer efficacement plusieurs systèmes entre eux, tout en assurant une bonne précision de l'estimation. La méthodologie repose sur une phase de caractérisation de composants IP matériels ainsi que de leur modélisation en Systeme. Dans un second temps, une représentation haut-niveau du système entier est réalisée à partir de la librairie des modèles Systeme de chaque IP. A travers des simulations haut-niveau, les utilisateurs peuvent tester rapidement de multiples configurations de leur système. Un des caractères innovants de l'approche repose sur l'utilisation de signaux clés qui permettent de tenir compte des comportements dynamiques des composants IP, c-à-d leur temps d'activité (actif/inactif), au sein du système et ainsi obtenir des estimations précises. Les nombreux gains de la méthodologie ont été démontrés à travers plusieurs exemples de systèmes de communications numériques sans fil comme une chaîne de traitement en bande de base de type SISO-OFDM générique, des émetteurs LTE etc. Pour conclure, les limitations ont été adressées et des solutions d'optimisation ont pu être envisagées puis mises en place. / Wireless communication systems are still evolving since the last decades, driven by the growing demand of the electronic market for energy efficient and high performance devices. Thereby, new design constraints have appeared that aim at taking into account power consumption in order to improve battery-life of circuits. Current wireless communication systems commonly dissipate a lot of power. On the other hand, the complexity of such systems keeps on increasing through the generations to always satisfy more users at a high degree of performance. In this highly constrained context, FPGA devices seem to be an attractive technology, able to support complex systems thanks to their important number of resources. According to the FPGA nature, designers need to estimate the power consumption and the performance of their wireless communication systems as soon as possible in the design flow. In this way, they will be able to perform efficient design space exploration and make decisive implementation and optimization choices. Throughout this thesis, a power estimation methodology for hardware-focused FPGA device is described and aims at making design space exploration a lot easier, providing early and fast power and performance estimation at high-level. It also proposes an efficient way to efficiently compare several systems. The methodology is effective through an lP characterisation step and the development of their SystemC models. Then, a high level description of the entire system is realized from the SystemC models that have been previously developed. High-level simulations enable to check the functionality and evaluate the power and performance of the system. One of the contributions consists in monitoring the JP time-activities during the simulation. We show that this has an important impact on both power and performances. The effectiveness of the methodology has been demonstrated throughout several baseband processing chains of the wireless communication domain such as a SISO-OFDM generic chain, LTE transmitters etc. To conclude, the main limitations of the proposed methodology have been investigated and addressed.

In today’s technologically driven world, antennas play an essential role in enabling wireless communications over long distances and allow communities to interact on a global scale. Typically, this is done using large networks of antenna systems on the ground and in space to ensure signals reliably arrive at their destinations, which can be very expensive in terms of replacement cost and maintenance as the number of nodes increase. We shall be discussing a new method for deploying space-borne antennas via CubeSats that will enable high speed communications with the ground and other satellites at a fraction of the cost and complexity of traditional communication systems. Specifically, the focus will be on the design of a phased array line feed and the backend systems for a CubeSat deployable inflatable spherical reflector that is electronically steerable to any direction in azimuth and elevation.

Det blir allt vanligare med att använda webbläsare som plattform för applikationer istället för de konventionella som måste installeras lokalt. Frågan är hur pass bra JavaScript står sig när det kommer till intensiva och tunga algoritmer såsom sökalgoritmer för spel. Finns det optimeringar som kan förbättra tidsåtgången för sökalgoritmer så att användaren i slutändan inte tröttnar på att det tar för lång tid? Med hjälp av ett antal tekniker såsom Typed Arrays försöker det här arbetet påvisa huruvida Typed Arrays kan användas som optimeringsstrategi för en applikation handskriven i JavaScript för att sänka tidsåtgången. Detta arbete syftar till att utvärdera hur användbart av optimeringsstrategier såsom Typed Arrays påverkar tidsåtgången för en förutbestämd sökväg. Flera versioner av en applikation implementeras för att se vilken optimering som är mest lämpad och utvärderas sedermera genom prestandamätning. Resultatet visar att ingen av de valda optimeringsstrategierna gav någon nämnvärd förbättring utan var till vissa delar en försämring jämfört med referensmätningen.

The existence of cosmic strings was proposed in the mid-seventies as a by-product of the various phase transitions that occured in the early Universe. Cosmic strings are one-dimensional topological defects; structures of extremely high energy density with infinitesimal widths and lengths of cosmological size. After they were proposed, cosmic strings with GUT energy scales became very popular as a potential source for galaxy formation, but after CMB observations ruled out this possibility, they stopped attracting much scientific attention. The whole field was revived as part of superstring theory, where the formation of cosmic (super)string networks is a very common characteristic of brane inflation models, allowing them to acquire energies over a much more extended range. Attempts to detect cosmic strings centers on the three most basic observational signatures they create: CMB anisotropies, gravitational lensing events and the stochastic gravitational wave background they are expected to have created. So far, no detection of cosmic strings has been achieved. Their non-detection has inevitably led to setting constraints on their most important characteristic; their lineal energy density (or tension) which describes their energy scale. The topic of this thesis is how to use pulsar timing arrays (PTAs) in order to set constraints on the string tension. The limits PTAs can set on the amplitude of the stochastic gravitational wave background at ~nHz frequencies can be used to set constraints on the string tension. Such an effort is much more complicated than CMB or gravitational lensing investigations due to the large number of unknown cosmic string model parameters which are involved and for which, not only we do not have any observational evidence for their value, but moreover, they can acquire values over very wide ranges. So far, previous investigations were based on assumptions about these parameters and on the specific gravitational wave emission mechanism from cosmic string loops. In this work we have constructed a new code to reproduce the gravitational wave background from a cosmic string network, based on the widely accepted one scale model. Using this, we have performed numerous simulations to study the effects on the gravitational wave spectrum for each cosmic string model parameter, covering the whole parameter space of interest for each of them. Moreover, we have also extended the application of our code in order to describe cosmic string networks which create loops on more than one scale, models of which have recently appeared in the literature. In particular, we have investigated cosmic string networks which create loops at two distinct scales and loops with scales described by a log-normal distribution After studying the properties of the gravitational wave spectrum from cosmic strings, we combined our simulations with the most stringent limit so far on the stochastic gravitational wave background imposed by the EPTA. This limit is provided as a function of the slope of the gravitational wave background and we have also used this information for the first time to acquire even more accurate results. In our approach, we did not make any assumption about the values of the cosmic string model parameters, investigating all possibilities and we managed to compute a conservative and completely general constraint on the cosmic string tension, G mu<5.3x10 -7, which is slightly weaker than the current constraints set by CMB and gravitational lensing. We concluded our work by estimating the projected constraints that are expected to be achieved by near future experiments like LEAP, and ultimately by the SKA, to find an improvement of at least two orders of magnitude, significantly outperforming the expected constraints by future CMB investigations.

De nos jours, les communications sans fil sont devenues un moyen incontournable et universel d'échange d'un large éventail d'informations entre différents systèmes, certains d'entre eux étant en mouvement comme des drones parmi tant d'autres. Dans le contexte de cette thèse nous considérons une liaison entre un projectile et une station de base. La géométrie de la structure ainsi que les contraintes aérodynamiques d'un tir balistique impliquent l'utilisation d'antennes patchs dans la partie conique à l'avant du projectile. Ce type d'antenne est facile à intégrer à une plate-forme en tant que réseau conformé tout en respectant les contraintes d'encombrement. Ces communications doivent être fiables et discrètes dans un environnement perturbé ou hostile. Les diagrammes de rayonnement du réseau d'antennes doivent présenter des caractéristiques spécifiques, notamment dans le cas particulier d'objets volants et possédant une rotation en roulis (rotation autour de son axe) qui impliquent l'utilisation d'un réseau phasé et commuté par rapport à sa position. Une antenne qui présenterait un rayonnement fixe assurerait une liaison avec un interlocuteur, mais rayonnerait également dans d'autres directions sensibles ce qui pourraient interférer avec la communication principale. La solution qui consiste à activer et désactiver des sous-réseaux verticaux afin d'orienter le lobe principal dans la plan orthogonal à la pointe semble être en accord avec les contraintes de la structure tournante. Un réseau conique a été étudié puis 2 prototypes ont été fabriqués, dont un à l'ISL. Les sous-réseaux sont répartis de manière égale autour de la pointe de façon à pouvoir rayonner dans toutes les directions. De plus, chaque sous-réseau est composé de trois éléments ce qui permet d'orienter également le lobe principal dans le plan longitudinal de la pointe (le long de l'axe du projectile) grâce à un dépointage électronique. Un système électronique de formation de faisceaux a été développé dans le but de contrôler 12 éléments rayonnants. Le réseau d'antennes ainsi que le circuit de répartition ont été caractérisés dans un premier temps de manière indépendante afin d'optimiser les lois de phase nécessaires à dépointer le lobe à partir des pondérations mesurées. Au final, le réseau de 12 éléments associé à son système d'alimentation dédié a été mesuré dans les chambres anéchoïques de DGA-MI et de l'ISL et les mesures sont en accord avec les simulations. / Nowadays wireless communications have become a useful and universal mean to exchange a wide range of information between different systems, some of them being moving, as UAVs among others. In this context we consider here the link between a projectile and a base station. The shape of the structure and the aerodynamic constraints involve the use of patch antennas in the conical front part. This class of antenna is easy to be integrated into the platform as a conformal array, while respecting space constraint. Communications have to be reliable and discrete in disturbed or hostile environment. Antennas array radiation patterns must have some specific characteristics, in particular in the case of flying objects with spin which involves the use of a switched phased array considering its roll position. A fixed-radiation pattern antenna may presents a relevant level or gain toward the interlocutor, but also toward sensitive directions, in which may be located others systems, interfering with the current communication. The solution to switch on and off vertical sub-arrays to steer the beam in the azimuthal plane seem convenient ant fitting the requirements of rotating platform. A conical phased array was studied and two prototypes were manufactured, one at ISL. Sub-arrays are distributed around the conical shape in order to be able to radiate in each direction. Moreover, each sub-array are composed of three radiating elements allowing to steer the main antenna beam in many direction (along the projectile fuze axes). A beam forming network was developed to control the 12 radiating elements conical array. The antenna array and the feeding network were characterized independently in order to optimized the phase of each radiating element. Finally, measurements were done on the whole system in the DGA-MI and ISL anechoic chambers and are in good agreement with simulation results.

L'objectif de cette thèse est d'exploiter les phénomènes des ondes de fuite pour améliorer les performances des antennes classiques à ouverture pour les applications spatiales. Ici, nous considérons des configurations planaires où les ondes de fuite sont excitées entre un plan de masse et un superstrat partiellement réfléchissant. Des réseaux de petites ouvertures sur le plan de masse sont utilisés pour alimenter les antennes considérées. Les structures avec superstrat sont développées en configuration réseaux simples ou réseaux à commande de phase, considérées intéressantes en termes de flexibilité du système pour des liens satellitaires de nouvelle génération.Pour étudier efficacement ces antennes, nous avons développé un outil d'analyse basé sur une approche spectrale avec fonction de Green. Cet outil permet d'analyser les structures proposées en prenant en compte l'impact du couplage mutuel entre les éléments sur les performances de rayonnement avec une réduction du temps de calcul et d'utilisation de mémoire.L'augmentation du gain des structures à ondes de fuite peut ouvrir la voie à la réduction du nombre d'éléments des réseaux associés, et donc des commandes en amplitude et phase. Dans une configuration à ondes de fuite, chaque élément du réseau rayonne avec une ouverture équivalente plus grande, augmentant ainsi l'espacement entre les éléments sans affecter le gain global de la structure. Comme largement expliqué dans le manuscrit, les solutions à ondes de fuite représentent par conséquent un avantage majeur pour les antennes du segment utilisateur.En plus de l'amélioration du gain, la technologie à ondes de fuite peut être efficacement exploitée pour synthétiser le diagramme de rayonnement, en choisissant correctement les paramètres de conception de l'antenne. Cette caractéristique peut être utilisée dans les réseaux à commande de phase, pour produire un diagramme d'élément qui minimise les pertes par dépointage et qui filtre les lobes de réseaux. Une procédure de synthèse pour des antennes réseaux raréfiés à ondes de fuite est ainsi présentée dans le manuscrit, ainsi qu'une nouvelle configuration de réseaux avec superstrat irrégulier. Ce dernier permet de réduire les lobes secondaires de l'antenne en utilisant une excitation uniforme. Cette dernière configuration montre clairement que la capacité de modifier le diagramme de rayonnement est la caractéristique la plus attrayante des antennes planaires à ondes de fuite, pour être utilisée dans des solutions de réseaux à commande de phase. / The objective of this thesis is to exploit the leaky-wave phenomena to enhance the performance of classical aperture antennas for space applications. Here, we consider planar configurations where the leaky modes are excited between a ground plane and a partially reflective superstrate. Arrangements of small apertures opening on the ground plane are used to feed the antennas under study. The superstrate-like leaky-wave structures are developed in array or phased array configurations, considered of interest in terms of flexibility of the system for next generation satellite links. In order to efficiently study planar leaky-wave arrays, we have developed an analysis tool based on a Green's function spectral approach. The developed tool allows to precisely analyze the proposed structure by taking into account the impact of the mutual coupling among the elements on the radiation performance of the whole antenna. In addition, it can handle extremely large structures in terms of wavelengths with a small computational effort with respect to commercial tools. In particular, the gain enhancement of leaky-based structures can pave the way to the reduction of the number of elements of the associated phased arrays. In a leaky-wave configuration each element of the array will radiate with a larger equivalent aperture allowing a larger spacing among elements without affecting the final gain of the whole structure. This aspect is particularly important in the case of phased arrays, where phase shifters and control cells are, typically, the most expensive components of the system. As extensively explained in the manuscript, antennas for user segment might find the highest benefit by using leaky-wave solutions. Besides the gain enhancement, the leaky-wave technology can be effectively exploited to conveniently shape the radiation pattern by properly engineering the design parameters of the antenna. This capability can be used in phased arrays to generate a convenient element pattern to minimize the scan losses and filter the grating lobes appearing in the visible space when dealing with periodicities larger than a wavelength. Therefore, a synthesis procedure for thinned leaky-wave arrays is presented in the manuscript. Also, a novel array configuration, the irregular superstrate array, is presented. The irregular superstrate allows the reduction of the side lobes of the antenna below -20 dB in the considered 2.5 % band, using a uniform excitation. This last configuration clearly shows that the shaping capability of leaky-wave antennas is the most appealing feature to be used in phased array solutions.

The principal contributions of this study include the development of a new capacitive feeding mechanism for wideband probe-fed microstrip patch antennas as well as the implementation of a spectral-domain moment-method formulation for the efficient analysis of large, but finite arrays of these elements. Such antenna configurations are very useful in the wireless communications industry, but extremely difficult to analyse with commercially available software. Probe-fed microstrip patch antennas have always been a popular candidate for a variety of antenna systems. Due to their many salient features, they are well suited for modern wireless communication systems. However, these systems often require antennas with wideband properties, while an inherent limitation of probe-fed microstrip patch antennas is its narrow impedance bandwidth. This can be overcome by manufacturing the antenna on a thick low-loss substrate, but at the same time it also complicates things by rendering the input impedance of the antenna very inductive. In this thesis, a new capacitive feeding mechanism is introduced that can be used for probe-fed microstrip patch antennas on thick substrates. It consists of a small probe-fed capacitor patch that is situated next to the resonant patch. The benefits of this configuration include the fact that only one substrate layer is required to support the antenna. It is also very easy to design and optimise. The use of full-wave methods for an accurate analysis of microstrip antennas, has basically become standard practice. These methods can become very demanding in terms of computational resources, especially when large antenna arrays have to be analysed. As such, this thesis includes a spectral-domain moment-method formulation, which was developed for the analysis of probe-fed microstrip patch antennas or antenna arrays that comprise of the new capacitive feeding mechanism. Here, entire-domain and subdomain basis functions are combined in a unique way so as to minimise the computational requirements, most notably computer memory. It is shown that, for general antenna array configurations, memory savings of more than 2500 times can be achieved when compared with typical commercial software packages where only subdomain basis functions are used. Some of the numerical complexities that are dealt with, include various methods to evaluate the spectral integrals as well as special algorithms to eliminate the recalculation of duplicate interactions. The thesis also contains a quantitative comparison of various attachment modes that are often used in the moment-method modelling of probe-to-patch transitions. Various numerical and experimental results are included in order to verify the spectral-domain moment-method formulation, to characterise the new feeding mechanism and to illustrate its use for various applications. These results show that, in terms of accuracy, the spectral-domain moment-method formulation compares well with commercial codes, while by comparison, it demands very little computer memory. The characterisation results show that the input impedance of the antenna can be fully controlled by only adjusting the size of the capacitor patch as well as the width of the gap between the capacitor patch and the resonant patch. In terms of applications, it is shown how the new antenna element can effectively be employed in linear arrays with vertical polarisation, horizontal polarisation or dual slant-polarisation. These represent some widely-used configurations for modern base-station antennas. / Thesis (PhD (Electronic Engineering))--University of Pretoria, 2005. / Electrical, Electronic and Computer Engineering / unrestricted

FPGA clock frequencies are slow enough that only a fraction of the interconnect’s bandwidth is used. By exploiting this bandwidth, the transfer of large amounts of data can be greatly accelerated. Alternatively, it may also be possible to save area on fixed-bandwidth links by using on-chip serial signaling. For datapath-intensive designs which operate on words instead of bits, this can reduce wiring congestion as well. This thesis proposes relatively simple circuit-level modifications to FPGA interconnect to enable high-bandwidth communication. High-level area estimates indicate a potential interconnect area savings of 10 to 60% when serial links are used. Two interconnect pipelining techniques, wave pipelining and surfing, are adapted to FPGAs and compared against each other and against regular FPGA interconnect in terms of throughput, reliability, area, power, and latency. Source-synchronous signaling is used to achieve high data rates with simple receiver design. Statistical models for high-frequency power supply noise are developed and used to estimate the probability of error of wave pipelined and surfing links as a function of link length and operating speed. Surfing is generally found to be more reliable and less sensitive to noise than wave pipelining. Simulation results in a 65nm process demonstrate a throughput of 3Gbps per wire across a 50-stage, 25mm link. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

The effectiveness of implementing field programmable gate arrays (FPGAs) in communication, military, space and high radiation environment applications, coupled with the increased accessibility of private individuals and researchers to launch satellites, has led to an increased interest in commercial off the shelf components. The metal oxide semiconductor (MOS) structures of FPGAs however, are sensitive to radiation effects which can lead to decreased reliability of the device. In order to successfully implement a FPGA based system in a radiation environment, such as on-board a satellite, the single event upset (SEU) and total ionizing dose (TID) characteristics of the device must first be established. This research experimentally determines a research procedure which could accurately determine the SEU cross sections and TID characteristics of various mitigation techniques as well as control circuits implemented in a ProASIC3 A3P1000 FPGA. To gain an understanding of the SEU effects of the implemented circuits, the test FPGA was irradiated by a 66MeV proton beam at the iTemba LABS facility. Through means of irradiation, the SEU cross section of various communication, motor control and mitigation schemes circuits, induced by high energy proton strikes was investigated. The implementation of a full global triple modular redundancy (TMR) and a combination of TMR and a AND-OR multiplexer filter was found to most effectively mitigate SEUs in comparison to the other techniques. When comparing the communication and motor control circuits, the high frequency I2C and SPI circuits experienced a higher number of upsets when compared to a low frequency servo motor control circuit. To gain a better understanding of the absorbed dose effects, experimental TID testing was conducted by irradiating the test FPGA with a cobalt-60 (Co-60) source. An accumulated absorbed dose resulted in the fluctuation of the device supply current and operating voltages as well as resulted in output errors. The TMR and TMR filtering combination mitigation techniques again were found to be the most effective methods of mitigation.