A wireless sensor network (WSN) is a collection of sensor nodes distributed over a geographical region to obtain the environmental data. It can have different types of applications ranging from low data rate event driven and monitoring applications to high data rate real time industry and military applications. Energy efficiency and reliability are the two major design issues which should be handled efficiently at all the layers of communication protocol stack, due to resource constraint sensor nodes and erroneous nature of wireless channel respectively. Media access control (MAC) is the protocol which deals with the problem of packet collision due to simultaneous transmissions by more than one neighboring sensor nodes. Time Division Multiple Access based (TDMA-based) and contention-based are the two major types of MAC protocols used in WSNs. In general, the TDMA-based channel access mechanisms perform better than the contention-based channel access mechanisms, in terms of channel utilization, reliability and power consumption, specially for high data rate applications in wireless sensor networks (WSNs). TDMA-based channel access employs a predefined schedule so that the nodes can transmit at their allotted time slots. Based on the frequency of scheduling requirement, the existing distributed TDMA-scheduling techniques can be classified as either static or dynamic. The primary purpose of static TDMA-scheduling algorithms is to improve the channel utilization by generating a schedule of smaller length. But, they usually take longer time to generate such a schedule, and hence, are not suitable for WSNs, in which the network topology changes dynamically. On the other hand, dynamic TDMA-scheduling algorithms generate a schedule quickly, but they are not efficient in terms of generated schedule length. We suggest a new approach to TDMA-scheduling for WSNs, that can bridge the gap between these two extreme types of TDMA-scheduling techniques, by providing the flexibility to trade-off between the schedule length and the time required to generate the schedule, as per the requirements of the underlying applications and channel conditions. The suggested TDMA-scheduling works in two phases. In the first phase, we generate a valid TDMA schedule quickly, which need not have to be very efficient in terms of schedule length. In the second phase, we iteratively reduce the schedule length in a manner, such that the process of schedule length reduction can be terminated after the execution of an arbitrary number of iterations, and still be left with a valid schedule. This step provides the flexibility to trade-off the schedule length with the time required to generate the schedule. In the first phase of above TDMA-scheduling approach, we propose two randomized, distributed and parallel TDMA-scheduling algorithms viz., Distributed TDMA Slot Scheduling (DTSS) and Randomized and Distributed TDMA (RD-TDMA) scheduling algorithm. Both the algorithms are based on graph coloring approach, which generate a TDMA schedule quickly with a fixed schedule length ( Colouring), where is the maximum degree of any node in the graph to be colored. The two algorithms differ in the channel access mechanism used by them to transmit control messages, and in the generated schedule for different modes of communication, i.e., unicast, multicast and broadcast. The novelty of the proposed algorithms lies in the methods, by which an uncolored node detects that the slot picked by it is different from the slots picked by all the neighboring nodes, and the selection of probabilities with which the available slots can be picked up. Furthermore, to achieve faster convergence we introduce the idea of dynamic slot-probability update as per which the nodes update their slot-probability by considering the current slot-probability of their neighboring nodes. Under the second phase of the proposed TDMA-scheduling approach, we provide two randomized and distributed schedule compaction algorithms, viz., Distributed Schedule Compaction (DSC) and Distributed Schedule Length Reduction (DSLR) algorithm, as the mechanism to trade-off the scheduling time with the generated schedule length. These algorithms start with a valid TDMA schedule and progressively compress it in each round of execution. Additionally, Furthermore, the execution of these algorithms can be stopped after an arbitrary number of rounds as per the requirements of underlying applications. Even though TDMA-based MAC protocols avoid packet loss due to collision, due to erroneous nature of wireless medium, they alone are not sufficient to ensure the reliable transmission in WSNs. Automatic Repeat reQuest (ARQ) is the technique commonly used to provide error control for unicast data transmission. Unfortunately, ARQ mechanisms cannot be used for reliable multicast/broadcast transmission in WSNs. To solve this issue, we propose a virtual token-based channel access and feedback protocol (VTCAF) for link level reliable multicasting in single-hop wireless networks. The VTCAF protocol introduces a virtual (implicit) token passing mechanism based on carrier sensing to avoid the collision between feedback messages. The delay performance is improved in VTCAF protocol by reducing the number of feedback messages. Besides, the VTCAF protocol is parametric in nature and can easily trade-off reliability with the delay as per the requirements of the underlying applications. Finally, by integrating all the works, viz., TDMA-scheduling algorithms (DTSS/RD-TDMA), schedule compaction algorithms and link layer feedback mechanism for reliable multicast/ broadcast, we propose a TDMA-based energy aware and reliable MAC protocol, named TEA-MAC for multi-hop WSNs. Similar to VTCAF, TEA-MAC protocol uses the combination of ACK-based and NACK-based approaches to ensure reliable communication. But, instead of using virtual token-based channel access, it uses contention-based channel access for NACK transmission. All the algorithms and protocols proposed in this thesis are distributed, parallel and fault tolerant against packet losses to support scalability, faster execution and robustness respectively. The simulations have been performed using Castalia network simulator to evaluate the performance of proposed algorithms/protocols and also to compare their performance with the existing algorithms/protocols. We have also performed theoretical analysis of these algorithms/protocols to evaluate their performance. Additionally, we have shown the correctness of proposed algorithms/protocols by providing the necessary proofs, whenever it was required. The simulation results together with theoretical analysis show that, in addition to the advantage of trading the runtime with schedule length, the proposed TDMA scheduling approach achieves better runtime and schedule length performance than existing algorithms. Additionally, the TEA-MAC protocol is able to considerably improve the reliability and delay performance of multicast communication in WSNs.

L’alternateur étudié dans cette thèse équipe une éolienne flottante à axe vertical. Il s’agit d’unemachine synchrone de 2MW, polyphasée, bobinée sur plots, à pas fractionnaire, à aimants et à prisedirecte. L’objectif majeur de la thèse est la caractérisation de cet alternateur afin d’en évaluer lesperformances. Une modélisation analytique de la perméance d'entrefer et de la force magnétomotriceont conduit à l'étude de l'induction d'entrefer dont le contenu harmonique s'est révélé êtreparticulièrement riche. La caractérisation a également porté sur la détermination des inductancespropres et mutuelles, avec un calage du modèle effectué en exploitant des mesures expérimentalesfaites à l’arrêt. La particularité de la machine étudiée nous a amené à considérer un enroulement fictiféquivalent dans le but de déterminer les inductances cycliques, nécessaires à l'établissement du schémamonophasé équivalent. De plus, le fonctionnement back to back de la machine a été étudié et simulé. Ilpermet de faire des tests de la machine elle-même mais aussi des éléments de la chaîne de conversion,sans l'utilisation d'une deuxième machine ou des pâles pour entraîner le rotor du prototype. Il offre enplus la possibilité de tester différents type de fonctionnements et d'évaluer la puissance que peutproduire la machine : différents scénario de fonctionnement ont été simulés et les paramètresnécessaires aux essais sur site ont été définis. Le fonctionnement back to back a fait l’objetd’investigation sur une machine synchrone classique. Les difficultés de fonctionnement ontnotamment pu être mises en évidence avec une modélisation analytique des phénomènesélectromagnétique qui s'y produisent et des essais expérimentaux. / The studied alternator equips a vertical axis offshore wind turbine. It is a 2MW fractional slotconcentrated winding permanent magnet synchronous polyphase machine, directly connected to bladeturbine. The principal purpose of this thesis consists in characterizing the alternator in order toevaluate its performances. The air gap permeance and the magnemotive force analytical modelingleads to study the air gap flux density and its harmonic content, which is particularly rich. Thecharacterization has also concerned the self and mutual inductance determination, which has requiredstalling the model by exploiting experimental measurements done at standstill. The particularity of thestudied machine leads us to consider an equivalent fictitious winding in order to determine the cyclicalinductances, necessary for the single-phase equivalent scheme establishment. Moreover, the machineback to back functioning has been studied and simulated. This functioning allows testing the machineitself and the other conversion chain subsystems, without using a second machine or blades to drivethe prototype rotor. It offers the possibility to test different types of operating points and to estimatethe power that can produce the studied machine: different operating points scenarios have beensimulated and parameters for tests have been defined. The back to back functioning of a classicalsynchronous machine has been also investigated, with an analytical modeling of the electromagneticphenomena and experimental tests.

This work deals with the most famous constructions of planar antennas. The work analyses selected methods of connecting planar antennas' charging with the possibility of impedance adaptation. An important part of the work consists of description of techniques for increasing bandwidth and conversely techniques for reducing the dimensions which are important for the design of planar antennas for mobile devices. The antennas were designed by means of theoretical basis and their properties were examined in the electromagnetic field simulator. The selected antenna model was constructed and its parameters were measured experimentally. The matching circuit for the antenna to tune it to different resonant frequencies was designed and constructed.

Disertační práce se zabývá návrhem a optimalizací kruhově polarizované anténa pro oblast terahertzových kmitočtů. V práci se věnuji zjednodušené teorii terahertzového zdroje a návrhu vhodné antény pro tento zdroj. Návrh je zaměřen na dosažení kruhové polarizace z lineárně polarizovaných antén. Abych potlačil šíření povrchové vlny na elektricky tlustém dielektrickém substrátu, věnuji se návrhu a optimalizaci specifických periodických struktur. Návrh těchto struktur je poměrně komplikovaný, protože neexistuje přímočarý vztah mezi vlastnostmi struktur s elektromagnetickým zádržným pásmem (EBG) a geometrií buňky. Abych vhodně koncentroval vyzařovanou energii do úzkého svazku, věnuji se návrhu a optimalizaci částečně odrazného plochy (PRS), které působí jako planární čočka pro terahertzovou anténu.

Les réseaux de capteurs acoustiques sous-marins (UW-ASN) sont les plus nouveaux achèvements technologiques en termes de communication. Les UW-ASN visent à observer et à explorer les lacs, les rivières, les mers et les océans. Récemment, ils ont été soumis à une attention particulière en raison de leur grand potentiel en termes d'applications prometteuses dans divers domaines (militaires, environnementaux, scientifiques ...) et aux nouvelles questions scientifiques qu'ils suscitent. Un problème majeur dans les UW-ASN est l'épuisement rapide de l'énergie, car une grande puissance est nécessaire pour la communication acoustique, tandis que le budget de la batterie des capteurs est limité. Par conséquent, les protocoles de communication énergétiques revêtent une importance primordiale pour faire usage judiciaire du budget énergétique disponible. Dans ce contexte, cette thèse vise à étudier les principales caractéristiques des capteurs acoustiques sous-marins difficiles afin de concevoir des protocoles de communication énergétiques, plus spécifiquement au niveau routage et MAC. Tout d'abord, nous abordons le problème des trous énergétiques dans UW-ASN. Le problème du « sink-hole » se produit lorsque les capteurs les plus proches du sink épuisent leur énergie plus rapidement en raison de leur charge plus lourde. En effet, ces capteurs, en particulier ceux qui sont à un seul saut du sinkstatique, agissent comme des relais pour tous les autres capteurs, ce qui leur épuise sévèrement l’énergie.A la couche de routage,en particulier, nous proposons de distribuer la charge transmise par chaque capteur parmi plusieurs voisins potentiels, en supposant que les capteurs peuvent ajuster leur gamme de communication entre deux niveaux lorsqu'ils envoient ou transmettent des données. Plus précisément, nous déterminons pour chaque capteur l'ensemble des prochains sauts avec les poids de charge associés qui entraînent un épuisement équitable d'énergie entre tous les capteurs du réseau. Ensuite, nous étendons notre stratégie de routage équilibrée en supposant que chaque capteur n'est pas seulement capable d'ajuster sa puissance d'émission à 2 niveaux mais aussi jusqu'à n niveaux où n> 2. Par conséquent, à la couche de routage, pour chaque valeur possible de n, nous déterminons pour chaque capteur l'ensemble des éventuels sauts avec les poids de charge associés qui mènent à une consommation d'énergie équitable chez tous les capteurs du réseau. En outre, nous obtenons le nombre optimal de puissances de transmission n qui équilibre la consommation d'énergie de tous les capteurs pour chaque configuration de réseau. En plus de cela, il convient de souligner que notre protocole de routage étendu utilise un modèle de canal à variation de temps plus réaliste qui tient compte de la plupart des caractéristiques fondamentales de la propagation acoustique sous-marine. Les résultats analytiques montrent que notre protocole de routage assure une réduction importante de la consommation d’énergie. Deuxièmement, pour atténuer les impacts de collision spectaculaires gaspillant l’énergie, nous concevons un protocole MAC multicanal (MC-UWMAC) évitant les collisions pour les UW-ASNs. MC-UWMAC fonctionne avec un canal de contrôle (décomposé en créneaux de temps) et un ensemble de canaux de données à bande passante égale. Les créneaux du canal de contrôle sont dédiés à l’échange RTS / CTS permettant à une paire de capteurs communicants de s'accorder sur l'heure de début de la communication sur un canal de données pré-alloué. Dans cette thèse, nous proposons deux nouvelles procédures associées d'allocation des créneaux du canal de contrôle et d'attribution des canaux de données sans nécessiter de frais de négociation supplémentaires. En conséquence, chaque capteur peut initier l'échange RTS / CTS uniquement à son créneau assigné, calculé à l'aide d'une procédure d'allocation basée sur une partition virtuelle de grille de la zone de déploiement. (...) / UnderWaterAcoustic Sensor Networks (UW-ASNs) are the newest technological achievement in terms of communication. Composed of a set of communicating underwater sensors, UW-ASNs are intended to observe and explore lakes, rivers, seas and oceans. Recently, they have been subject to a special attention due to their great potential in terms of promising applications in various domains (military, environmental, scientific...) and to the new scientific issues they raise. A great challenging issue in UW-ASNs is the fast energy depletion since high power is needed for acoustic communication while sensors battery budget is limited. Hence, energy-efficient networking protocols are of a paramount importance to make judicious use of the available energy budget while considering the distinguishing underwater environment characteristics. In this context, this thesis aims at studying the main challenging underwater acoustic sensors characteristics to design energy-efficient communication protocols specifically at the routing and MAC layers. First, we address the problem of energy holes in UW-ASNs. The sink-hole problem occurs when the closest nodes to sink drain their energy faster due to their heavier load. Indeed, those sensors especially the ones that are 1-hop away from the static sink act as relays to it on behalf of all other sensors, thus suffering from severe energy depletion. In particular, at the routing layer, we propose to distribute the transmission load at each sensor among several potential neighbors, assuming that sensors can adjust their communication range among two levels when they send or forward data. Specifically, we determine for each sensor the set of next hops with the associated load weights that lead to a fair energy depletion among all sensors in the network. Then, we extend our balanced routing strategy by assuming that each sensor node is not only able to adjust its transmission power to 2 levels but eventually up to n levels where n > 2. Consequently, at the routing layer, for each possible value of n, we determine for each sensor the set of possible next hops with the associated load weights that lead to a fair energy consumption among all sensors in the network. Moreover, we derive the optimal number of transmission powers n that balances the energy consumption among all sensors for each network configuration. In addition to that, it is worth pointing out that our extended routing protocol uses a more realistic time varying channel model that takes into account most of the fundamental characteristics of the underwater acoustic propagation. Analytical results show that further energy saving is achieved by our extended routing scheme. Second, to mitigate the dramatic collision impacts, we design a collision avoidance energy efficient multichannel MAC protocol (MC-UWMAC) for UW-ASNs. MC-UWMAC operates on single slotted control and a set of equal-bandwidth data channels. Control channel slots are dedicated to RTS/CTS handshaking allowing a communicating node pair to agree on the start time of communication on a pre-allocated data channel. In this thesis, we propose two novel coupled slot assignment and data channels allocation procedures without requiring any extra negotiation overhead. Accordingly, each node can initiate RTS/CTS exchange only at its assigned slot calculated using a slot allocation procedure based on a grid virtual partition of the deployment area. Moreover, for each communicating pair of nodes, one data channel is allocated using a channel allocation procedure based on our newly designed concept of singleton- intersecting quorum. Accordingly, each pair of communicating nodes will have at their disposal a unique 2-hop conflict free data channel. Compared with existing MAC protocol, MC-UWMAC reduces experienced collisions and improves network throughput while minimizing energy consumption.

The growing need for simpler and cheaper manufacturing process has led to the research into fractional-slot concentrated-wound (FSCW) motors. This concept has been widely investigated for surface-mounted permanent magnet (SMPM) machines. This thesis studies the same concept applied for synchronous reluctance machines (SynRM). In this thesis, a FSCW, 15 kW, 4-pole, Permanent-Magnet-Assisted Synchronous Reluctance Machine (PMaSynRM) is designed and optimized using finite element method (FEM) based simulations for a set of given technical specifications. Initially, the existing synchronous machine topologies are investigated and later two novel motor designs are introduced and optimized, namely, a FSCW-SynRM and a FSCW-PMaSynRM with ferrite magnets. Moreover, the influence of replacing ferrite material with Neodymium Iron Boron (NdFeB) in the FSCWPMaSynRM is analyzed. Detailed investigations are performed in order to compare the impact of material at different temperatures. Variation of the torque-speed capabilities with temperature and a safe operating temperature range where the magnets are not demagnetized are identified. The variation of overload capability with temperature is also studied. Finally, a comparison between the new proposed designs and other existing standard design topologies is performed. It was found that FSCW-SynRM present lower efficiency, power factor and higher torque ripple than DW-SynRM. However when ferrite magnets are inserted in FSCW-PMaSynRM the efficiency, power factor and the flux-weakening capability exceed the values of the DW-SynRM. Moreover, by using NdFeB instead of ferrite in FSCW-PMaSynRM, the torque ripple, the fluxweakening capability and the overload capability improve and a wider safe temperature range for no demagnetization is achievable. Finally, it is found that DW-PMaSynRM with ferrite presents the same efficiency level as FSCW-PMaSynRM with ferrite, but higher power factor and lower torque ripple. However FSCW-PMaSynRM with ferrite have other advantages, such as shorter end-winding length, good fault-tolerant capability and simpler and cheaper manufacturing process. / Det växande behovet av enklare och billigare tillverkningsprocesser har lett till att den senaste forskningen om elektriska maskiner har fokuserats till maskiner med koncentrerade lindningar (FSCW). Detta koncept har i stor utsträckning undersökts för synkronmaskiner med ytmonterade permanentmagneter (SMPM). Detta projekt studerar samma lindning koncept, tillämpat på synkrona reluktansmaskiner (SynRM). I denna avhandling är en 15 kW, 4-polig, SynRM med FSCW och permanentmagneter (PMaSynRM) utformad och optimerad med användandet av finita elementmetoden (FEM). Simuleringar för en uppsättning givna, tekniska specifikationer har genomförts. Inledningsvis undersöks den befintliga synkronreluktansmaskinen med distribuerad lindning och senare presenteras och optimeras två nya motorkonstruktioner: en FSCW-SynRM respektive en FSCW-PMaSynRM med ferritmagneter. Vidare analyseras påverkan av att ersätta ferritmaterial med neodym-järn-bor (NdFeB) i FSCW-PMaSynRM. Detaljerade undersökningar genomförs för att jämföra effekten av materialen vid olika temperaturer. Variationen av maximal vridmomentet som funktion av hastighet vid olika temperaturer identifierar ett säkert driftstemperaturintervall där magneterna inte avmagnetiseras. Även variationen i överbelastningskapacitet vid olika temperaturer studeras. Slutligen görs en jämförelse mellan den nya föreslagna designen och andra befintliga topologier. Resultaten visar att FSCW-SynRM har lägre effektivitet och effektfaktor, samt högre vridmomentsrippel än DW-SynRM. När ferritmagneter är införda i FSCW-PMaSynRM erhålls emellertid högre värden på effektivitet, effektfaktor och fältförsvagning än i DW-SynRM. Genom att använda NdFeB i stället för ferrit i FSCW-PMaSynRM förbättras dessutom vridmomentet, fältförsvagningskapaciteten och överbelastningskapaciteten, vilket ger ett bredare temperaturområde utan avmagnetisering. Slutligen visar DW-PMaSynRM med ferrit samma effektivitetsnivå som FSCW-PMaSynRM med ferrit, men med högre effektfaktor och lägre vridmomentsrippel. FSCW-PMaSynRM med ferrit har dock andra fördelar, såsom kortare härvänderna, god feltolerans samt enklare och billigare tillverkningsprocess.

The emergence of Ultra-wideband (UWB) technology application has yielded tremendous and vital impacts in the field of microwave wireless communications. These applications include military radar imaging, security screening, and tumour detection, especially for early detection of breast cancer. These indicators have stimulated and inspired many researchers to make the best use of this promising technology. UWB technology challenges such as antenna design, the problem of imaging reconstruction techniques, challenges of severe signal attenuation and dispersion in high loss material. Others are lengthy computational time demand and large computer memory requirements are prevalent constraints that need to be tackled especially in a large scale and complex computational electromagnetic analysis. In this regard, it is necessary to find out recently developed optimisation techniques that can provide solutions to these problems. In this thesis, designing, optimisation, development, measurement, and analysis of UWB antennas for near-field microwave imaging applications are considered. This technology emulates the same concept of surface penetrating radar operating in various forms of the UWB spectrum. The initial design of UWB monopole antennas, including T-slots, rectangular slots, and hexagonal slots on a circular radiating patch, was explicitly implemented for medical imaging applications to cover the UWB frequency ranging from 3.1 GHz to 10.6 GHz. Based on this concept, a new bow-tie and Vivaldi UWB antennas were designed for a through-the-wall imaging application. The new antennas were designed to cover a spectrum on a lower frequency ranging from 1 GHz - 4 GHz to ease the high wall losses that will be encountered when using a higher frequency range and to guarantee deeper penetration of the electromagnetic wave. Finally, both simulated and calculated results of the designed, optimised antennas indicate excellent agreement with improved performance in terms of return loss, gain, radiation pattern, and fidelity over the entire UWB frequency. These breakthroughs provided reduced computational time and computer memory requirement for useful, efficient, reliable, and compact sensors for imaging applications, including security and breast cancer detection, thereby saving more lives. / Tertiary Education Trust Fund (TET Fund) Supported by the Nigerian Defence Academy (NDA)

The master´s thesis deal with the thesis of a single-phase induction motor that is more and more used all over the world. Its aim is assembling the information about this type of motor and its calculation. The content of the first part is the theoretical analysis of a single-phase induction motor and acquirement ways of the stroke moment. Than the work contains a project of basic schema of the single-phase induction motor with an auxiliary phase. The practical part is devoted to an elaboration of demonstration example of the single-phase induction motor with concrete parameters calculation, its method was projected before. Than there is practising successived measuring on the motor of the same parameters made firm EMP s.r.o. Slavkov by Brno in this part. In the conclusion there is the evaluation of the whole process with reached results.

In the first part of the master’s thesis dealing with the increasing efficiency of induction motors there are briefly presented basic information about induction motors, followed by an overview of the losses of induction motors. The next part deals with the ways to increase efficiency of induction motors without increasing tooling costs. The practical part consists of four measurements of four induction motors, with their various mechanical adjustments to make comparing benefits of these modifications possible. The measured results are compared by a finite element method in Maxwell 2D Design program, in which the same motors are simulated as measured. Theoretical knowledge about the increase of efficiency is practically applied while being implemented in the simulations.

Diploma Thesis „Polyfunctional building“. The work deals implementation project of polyfunctional building in Jihlava. Object is situated on Jihlava suburb in emerging part called „Horní Kosov“. Land is gently sloping from west to east. It continues for Buková street. The building has six floors and a basement. The basement is made from monolithic reinforced concrete. It is used as garages. Floors are made of ceramic bricks with isolation . First floor contains four commercial premises and main entry for housing part. Seventeen flats are situated from second to sixth floor. Sizes of apartments are 1+kitchen corner to 4+kitchen corner with terraces and balcony. Roofing is solved with simple flat roof. Heating will be ensured with central heating. Every apartment, commercial premises, shared premises are going to have it’s own gas boiler. Polyfunctional building is connected on a local road – Buková street, which is connected to new communication. Within multifunctional building twenty parking slots is designed at the basement. Next fourteen parking slots is designed outside.