Wireless networking is fast becoming the primary method for people to connect to the Internet and with each other. The available wireless spectrum is increasingly congested, with users demanding higher performance and reliability from their wireless connections. This thesis proposes a game-theoretic random access model, compliant with the IEEE 802.11 standard, which can be integrated into the distributed coordination function (DCF). The objective is to design a game theoretic model that potentially optimizes throughput and fairness in each node independently and, therefore, minimise channel access delay. This dissertation presents a game-theoretic MAC layer implementation for single-cell networks and centralised DCF in the presence of hidden terminals to show how game theory can be applied to improve wireless performance. A utility function is proposed, such that it can decouple the protocol's dynamic adaptation to channel load from collision detection. It is demonstrated that the proposed model can reach a Nash equilibrium that results in a relatively stable contention window, provided that a node adapts its behaviour to the idle rate of the broadcast channel, coupled with observation of its own transmission activity. This dissertation shows that the proposed game-theoretic model is capable of achieving much higher throughput than the standard IEEE 802.11 DCF with better short-time fairness and significant improvements in the channel access delay.

Applications are now starting to emerge for superconducting devices in the areas of electrical power conversion and management, for example superconducting windings for marine propulsion motors, superconducting fault current limiters and superconducting magnet energy storage (SMES). Many of these applications also require power electronics, and it is therefore timely to consider the possibility of locating the power electronics in the cryosystem with the superconducting devices. Although significant work has been undertaken on the cryogenic operation of small devices, little has been published on larger devices, particularly the IGBT. This therefore forms the focus of this study. To examine the cryogenic performance of the sample devices, a cryo-system consisting of a cold chamber, a helium-filled compressor and vacuum pumps was built. Static, gate charge and switching tests were carried out on three types of IGBT modules, PT (punch-through), NPT (non-punch-through) and IGBT3 respectively, in the temperature range of 50 to 300 K. The switching tests were undertaken at 600V and up to 110 A. A physically based, compact level-1 model was selected to model the cryogenic performance of the IGBTs. A generic Saber power diode model with reverse recovery was selected to model the diode cryogenic performance. Close correspondence was demonstrated between the models and experimental results over the temperature range of 50- 300 K. Saber simulation was used to examine the cryogenic performance of a DC-DC step-down converter and a pulse-width modulated inverter leg, in which the temperature-dependent power device models developed in the modelling work were used. The simulation results showed that standard power electronic circuits using standard devices could work much more efficiently at low temperatures, for example, the efficiency of the DC-DC converter working at 50 kHz being increased from 90.0% at room temperature to 97.0% at 50 K.

This thesis has theoretically and experimentally investigated the ultrashort optical pulse propagation in semiconductor optical amplifiers (SOAs), which is an important topic in the optical fiber communication and optical signal processing. Some new work has been done: Firstly, effects of carrier heating on the ultrashort optical pulse propagation in quantum well SOAs are first studied taking into account the holes’ non-parabolic density of states; for bulk SOAs, an accurate and simple analytical method to study carrier heating effects is presented based on Fermi-Dirac integrals approximation. Secondly, this thesis reports a novel bias current optimization method for ultrashort optical pulse distortionless amplification in SOAs based on the newly proposed bias current relation function. Detailed theoretical and experimental work is done to analyze the relation between the optimized bias current and the parameters of the input ultrashort pulse train. Finally, a novel modelling technique-quantum transmission line modelling (Q-TLM) method is proposed by combining quantum statistic description and photon-electron dynamic interaction process description. Q-TLM is used to establish models for quantum well and quantum dot structures and analyze the dynamic performance of ultrashort optical pulse propagation in SOAs. The Q-TLM technique provides an effective method to study semiconductor optical devices.

A novel hybrid HVDC system is proposed based on the traditional LCC HVDC system. The proposed system is able to achieve full elimination of commutation failures which cannot be achieved in traditional LCC HVDC systems. In addition, reactive power controller is designed for the hybrid HVDC system. The controller is able to achieve zero reactive power exchange with the connected AC system at inverter side. It can also facilitate a faster fault recovery. Finally, the black start capability of the hybrid system is investigated. The black start sequence and inverter AC voltage controller are designed to achieve smooth and reliable black start of inverter AC system. The performances of the proposed system and controller are validated through detailed simulations in Real Time Digital Simulator (RTDS).

This thesis explores the application of low-frequency electromagnetic fields, which may be excited within a buried pipe, for the detection of underground utilities. Low-cost network analyser technology, which can be applied to field-measurements of the relative-permittivity of soil, is evaluated. These technologies are compared to laboratory-grade alternatives whose cost prohibits their use for field work. Methodologies for the measurement of the relative-permittivity of soil are discussed with reference to the low-cost technology, including use of a novel coaxial cavity which incorporates a step-discontinuity. It is shown that there is potential for use of low-cost network analysers in measuring relative-permittivity, but that further research is required to formulate a complete methodology. The propagation of electromagnetic waves in layered media is discussed. The recent literature relating to this field is extensively reviewed, with several errors and omissions highlighted. A new calculation is presented which allows the calculation of the electromagnetic field due to a vertical electric dipole in a four-layered medium. Example results, including an approximation of a leaking pipe, are presented. Finally, two sets of field trials are reviewed. The first field trials looked to observe waves propagating with low-velocity in the ground, by measuring the phase change along an array of receiving probes. Waves, propagating with low-velocity, were observed. However, direction of arrival measurements were not achievable due to a combination of signal-to-noise ratio, and the expected phase change at the observed propagation-velocity, across an array of realistic size. The second field trials measured low-frequency electromagnetic fields, excited within a buried pipe, which were used to detect the location of the pipe with good correspondence to the ground truth. Furthermore, comparison with a ground-penetrating radar survey indicated that some anomalous results in the low-frequency electromagnetic survey corresponded to shallow targets detected using ground-penetrating radar.

This thesis describes a research into multi-frequency and filtering antennas. Several novel antennas are presented, each of which addresses a specific issue for future communication systems, in terms of multi-frequency operation, and filtering capability. These antennas seem to be good candidates for implementation in future multiband radios, cognitive radio (CR), and software defined radio (SDR). The filtering antenna provides an additional filtering action which greatly improves the noise performance and reduces the need for filtering circuitry in the RF front end. Two types of frequency reconfigurable antennas are presented. One is tunable left-handed loop over ground plane and the second is slot-fed reconfigurable patch. The operating frequency of the left handed loop is reconfigured by loading varactor diodes whilst the frequency agility in the patch is achieved by inserting switches in the coupling slot. The length of the slot is altered by activating the switches. Compact microstrip antennas with filtering capabilities are presented in this thesis. Two filtering antennas are presented. Whilst the first one consists of three edge-coupled patches, the second filtering antenna consists of rectangular patch coupled to two hairpin resonators. The proposed antennas combine radiating and filtering functions by providing good out of band gain suppression.

Microwave circuits in the millimetre wave region demand low loss, and low dispersion transmission lines. The work carried out in this thesis is on low loss transmission lines and filters, based on a square coaxial transmission line which is made only of metal, avoiding dielectric and radiation losses. The metal structure inside the square coaxial transmission line is supported by stubs, which provide the mechanical support for the centre conductor for the coaxial transmission lines and filters. The coaxial structure is made by stacking thick planar layers of material to suit microfabrication, providing the means to design high Q Microwave and RF passive devices, this transmission line structure is compact compared with a microstrip or a stripline, and gives better loss performance. Through this thesis, the way of optimising the square coaxial transmission line to provide a low loss will be presented, which will end in the presentation of one dielectric supported coaxial structure and three self supported filters, three of them were designed for the X-band, and one of them was designed for the Ka band. The application of the coaxial transmission line is demonstrated with wideband and narrow band designs.

This thesis presents the designs and measurement performance of nine SU-8 micromachined waveguide circuits operating at WR-10 band (75-110 GHz), WR-3 band (220-325 GHz) and WR-1.5 band (500-750 GHz). Two thick SU-8 photoresist micromachining processes, namely, the separate single-layer process and the joint two-layer process, are developed to fabricate these terahertz waveguide circuits. In order to achieve accurate and secure interconnections with measurement network analyzers, two calibrated measurement methods for micromachined waveguide circuits are proposed. The measurement performance of these micromachined circuits is excellent in terms of very low insertion loss. The design of multiple-passband filters using coupling matrix optimisation is also discussed in this thesis. The optimisation is performed on the coupling matrix and a genetic algorithm (GA) is employed to generate initial values for the control variables for a subsequent local optimisation (sequential quadratic programming - SQP search). The novel cost function presented in this thesis measures the difference of the frequency locations of reflection and transmission zeros between the response produced by the coupling matrix and the ideal response. An eighth-order X-band dual-band waveguide filter with all capacitive coupling irises is fabricated and measured to verify the design technique.

Independent component analysis (ICA) has the potential of automatically determining metabolite, macromolecular and lipid (MMLip) components that make up magnetic resonance (MR) spectra. However, the realiability with which this is accomplished and the optimal ICA approach for investigating in vivo MR spectra, have not yet been determined. A wavelet shrinkage de-noising based enhancement algorithm, utilising a newly derived relationship between the real and imaginary parts of the MR spectrum, is proposed. This algorithm is more robust compared with conventional de-noising methods. The two approaches for applying ICA, blind source separation (BSS) and feature extraction (FE), are thoroughly examined. A feature dimension selection method, which has not been adequately addressed, is proposed to set a theoretical guideline for ICA dimension reduction. Since the advantages and limitations of BSS-ICA and FE-ICA are different, combining them may compensate their disadvantages and lead to better results. A novel ICA approach involving a hybrid of the two techniques for automated decomposition of MRS dataset is proposed. It has been demonstrated that hybrid ICA provides more realistic individual metabolite and MMLip components than BSS-ICA or FE-ICA. It can aid metabolite identification and assignment, and has the potential for extracting biologically useful features and discovering biomarkers.

The advent of wireless location technology and the increase in location-based services, has meant the need to investigate efficient network-based location methods becoming of paramount importance. Therefore, the interest in wireless positioning techniques has been increasing over recent decades. Among mobile positioning techniques, the Time of Arrival (TOA) and Time Difference of Arrival (TDOA) look promising. For the purpose of dealing with such technologies, some classic algorithms such as least square, most likelihood and Taylor method have been used to solve the estimation, which distinguishes the location. However, in real practice, there are certain factors that influence the level of location accuracy. The two most significant factors are cellular topologies and non-line-of-sight (NLOS) effect. This thesis reviews existing approaches and suggests innovative methods for both line-of-sight (LOS) and NLOS scenarios. A simulation platform is designed to test and compare the performances of these algorithms. The results of the simulation compared with actual position measurements demonstrate that the innovative approaches have high positioning accuracy. Additionally, this thesis demonstrates different types of cellular topologies and develops a simulation to show how the cellular topology affects the positioning quality level. Finally, this thesis implements an experiment to exhibit how the innovative algorithms perform in the real world.