Over the past two decades quantum-dot-based photovoltaic devices have been attracting a lot of attention due to their potential high efficiencies and low cost fabrication. Unlike conventional photovoltaic devices where the absorption of a single photon always produces a single electron hole pair (exciton), quantum-dot-based devices can generate multiple excitons from the absorption of just a single photon. Thanks to this process, which is referred to as either carrier multiplication or multiple excition generation, quantum-dot-based devices can potentially reach higher efficiencies breaking the Shockley-Queisser limit. In addition, the colloidal synthesis techniques used to fabricate these devices are potentially very cheap and scalable. Despite the intrinsic potential of these devices, they are not currently at a stage where they can compete with commercial photovoltaics. In this thesis various factors that effect the efficiency of carrier multiplication are investigated. In addition new analytical methods are developed to form a contribution to theoretical work in this field.

There are many sources of waveforms or signals existing around us. They can be natural phenomena such as sound, light and invisible like electromagnetic fields, voltage, etc. Getting an insight into these waveforms helps explain the mysteries surrounding our world and the signal spectral analysis (i.e. the Fourier transform) is one of the most significant approaches to analyze a signal. Nevertheless, Fourier analysis cannot provide a time-dependent spectrum description for spectrum-varying signals-non-stationary signal. In these cases, time-frequency distribu- tions are employed instead of the traditional Fourier transform. There have been a variety of methods proposed to obtain the time-frequency representations (TFRs) such as the spectrogram or the Wigner-Ville distribution. The time-frequency distributions (TFDs), indeed, offer us a better signal interpretation in a two-dimensional time-frequency plane, which the Fourier transform fails to give. Nevertheless, in the case of incomplete data, the time-frequency displays are obscured by artifacts, and become highly noisy. Therefore, signal time-frequency features are hardly extracted, and cannot be used for further data processing. In this thesis, we propose two methods to deal with compressed observations. The first one applies compressive sensing with a novel chirp dictionary. This method assumes any windowed signal can be approximated by a sum of chirps, and then performs sparse reconstruction from windowed data in the time domain. A few improvements in computational complexity are also included. In the second method, fixed kernel as well as adaptive optimal kernels are used. This work is also based on the assumption that any windowed signal can be approximately represented by a sum of chirps. Since any chirp's auto-terms only occupy a certain area in the ambiguity domain, the kernel can be designed in a way to remove the other regions where auto-terms do not reside. In this manner, not only cross-terms but also missing samples’ artifact are mitigated significantly. The two proposed approaches bring about a better performance in the time-frequency signature estimations of the signals, which are sim- ulated with both synthetic and real signals. Notice that in this thesis, we only consider the non-stationary signals with frequency changing slowly with time. It is because the signals with rapidly varying frequency are not sparse in time-frequency domain and then the compressive sensing techniques or sparse reconstructions could not be applied. Also, the data with random missing samples are obtained by randomly choosing the samples’ positions and replacing these samples with zeros.

With today’s challenges in the electronic industry, novel alternative ap- proaches for manufacturing devices at nanoscale are being investigated. Using self-assembly, arguably has the best potential for nanostructures. DNA and proteins - some of the most important biomolecules use self- assembly extensively for natural functions. Chemical and structural pre- dictability of DNA and specificity of proteins promise a big potential for novel materials and could allow creation of structures controlled at nanoscale level. RecombinaseA - a DNA-binding protein has been used for controllable and predictable patterning of selected DNA sequences, opening the way to nanometre-scale DNA marking. However, protein patterning alone does not add any electric or other desired functionality to the DNA, there- fore additional modifications are neccessary. Furthermore, since biologi- cal molecules have transient functionality, system stability investigation is crucial for needed modification and subsequent usage. This project focused on RecA-patterned DNA modification for electric prop- erty addition. Thiolation and subsequent attachment of gold or magnetic nanoparticles to RecA protein present on DNA were investigated as a method for creating electrically conductive nanoscale objects. More specifically, at- tachment of gold nanoparticles throughout the whole patterned region of DNA and attachment of single nanoparticles at precise positions were looked into. The work successfully demonstrated that both nanoparticle deposition along the full length of RecA-coated DNA and specific single nanoparticle positioning is feasible. For investigating RecA-DNA stability, a system based on FRET was de- vised and used to analyse interaction kinetics. It was found that RecA-DNA complexes are fully formed in minutes and stay bound for hours. Specific configurations of the set-up showed distinct lack of signal, suggesting com- plicated interactions between the protein and patterned DNA. The project demonstrated through binding of NPs at specific locations and on the whole filament length that the system has potential for electronic applications and its stability is sufficient for processing times.

Nano-electromechanical systems (NEMS) are an important new class of device, with a growing range of applications, from tests of quantum mechanics through to nanoscale metrology and a vast number of different sensors. Cryogenic operation is also possible, and at low temperatures, nanoscale resonators exhibit quantum behaviour. NEMS resonators require readout of ultra-small, atomic scale displacements. To achieve this at low temperatures we have developed an ultrasensitive nanoSQUID readout of a coupled current-carrying NEMS resonator. The NanoSQUIDs are fabricated by gallium focussed ion beam milling and are based on niobium nanobridge weak links (Dayem bridges). The nanoSQUIDs have low loop inductance and low junction capacitance resulting in high flux and energy sensitivity. This work focusses on the characterisation of the resonator and nanoSQUID after they have been incorporated onto one chip. This is done through nanoindentation to characterise resonators and electronic measurements of the SQUID using a low-temperature preamplifier. It is found that the model used based upon an Euler-Bernoilli beam is correct close the centre of the sample but does not fit data points well close to the contacts. It is found the resonators have Young’s modulus in the range of 3GPa-241GPa. Both beam and paddle-shaped resonators are investigated and the models are made based upon the two different shapes. That for the paddle is based upon the same as the beam but uses a rectangular function to describe the changing area moment of inertia along the length of the resonator. The SQUID devices are characterised and found to have a typical noise floor of 0.2μ 0/pHz. Problems which have arisen due to the orientation of the two magnetic fields and their effect on the SQUID performance are discussed. We consider the geometry and optimum coupling of rectangular and square Si3N4 resonators to matching similar shaped nanoSQUID loops. We also discuss simulations of the nanoSQUID response versus resonator position for both symmetric and asymmetric configurations. It is found that optimal coupling is achieved in the asymmetric case due to the cancelling of the change in flux in the symmetric case. The use of a normal conducting or superconducting resonator is compared. It is found that a superconducting resonator provides a much larger SQUID response when actuated towards​ the device but cannot be used in the regime due to limitations of the superconducting transition temperature of Al (the resonator) being lower than the non-hysteretic operable temperature of the SQUID. Preliminary measurements are conducted on the coupled devices. It is noted that the signal from the device in the conducting case may be read out at 2! due to the sinusoidal change in flux through the SQUID loop and position of the resonator. The possibility of measuring such a signal is first investigated using a spectrum analyser but it is found the SQUID is pushed to nonlinear regions of its transfer curve. This results in a component of the signal at 2! due to the nonlinearity of the SQUID response. Conditions under which the SQUID is still operating in small signal mode (to preserve linearity of the SQUID response) are considered and from this we conclude there is a need to use phase sensitive detection to achieve optimum sensitivity. This technique is used to conduct the final measurement of the motion of the resonator by the SQUID and a preliminary result is found.

An increasing number of Low Latency Applications (LLAs) in the entertainment (Virtual/Augmented Reality), Internet-of-Things (IoT), and automotive domains require response times that challenge the traditional application provisioning process into distant data centres. At the same time, there is a trend in deploying In-Network Computing Resources (INCRs) closer to end users either in the form of network equipment, with capabilities of performing general-purpose computations, and/or in the form of commercial off-the-self “data centres in a box”, i.e., cloudlets, placed at different locations of Internet Service Providers (ISPs). That is, INCRs extend cloud computing at the edge and middle-tier locations of the network, providing significantly smaller response times than those achieved by the current “client-to-cloud” network model. In this thesis, we argue about the necessity of exploiting INCRs for application provisioning with the purpose of improving LLAs’ Quality of Service (QoS) by essentially deploying applications closer to end users. To this end, this thesis investigates the deployment of LLAs over INCRs under fixed, mobile, and disrupted user connectivity environments. In order to fully reap the benefits of INCRs, we develop for each connectivity scenario algorithmic frameworks that are centred around the concept of a market, where LLAs lease existing INCRs. The proposed frameworks take into account the particular characteristics of INCRs, such as their limited capacity in hosting application instances, and LLAs, by addressing the number of instances each application should deploy at each computing resource over time. Furthermore, since typically the smooth operation of network applications is supported by Network Functions, such as load balancers, firewalls etc., we consider the deployment of complementary Virtual Network Functions for backing LLAs’ provisioning over INCRs. Overall, the key goal of this thesis is the investigation of using an enhanced Internet through INCRs as the communication platform for LLAs.

This research work focuses on the design and characterisation of indium phosphide photonic integrated circuits for continuous-wave, high-purity millimetre-wave (mm-wave) and terahertz (THz) signal generation. Each source investigated in this work is based on a photonic oscillator (PO) consisting of monolithically integrated semiconductor lasers, and a broad bandwidth photodiode to convert the heterodyne signal from the optical to the electrical domain. The resulting photocurrent contains a component at the desired mm-wave or THz frequency corresponding to the frequency difference between the two lasers. In this thesis, the specifics of three POs are discussed, and dedicated laser phase-locking solutions are investigated and implemented, resulting in a high-purity mm-wave photonic synthesiser being realised and a novel THz PO being proposed and constructed. The former is a compact mm-wave photonic synthesiser consisting of two lasers monolithically integrated with fast photodiodes. High-quality, low-phase-noise signal above 100 GHz is demonstrated through optical injection locking, allowing the synthesised signal to be finely tuned across a 30 GHz span. Furthermore, the phase stabilisation scheme based on optical phase lock loop (OPLL) was constructed and discussed. The latter is a broadly tuneable THz signal generator based on a photonic integrated circuit developed using a generic fabrication foundry approach. The implementation of the photonic chip with twin OPLL enables the two optical lines to combine at the output and create a high-purity, continuously tuneable optical heterodyne signal, which can be data modulated. Furthermore, OPLL operation principles are investigated, leading to the establishment of design guidelines and a definition of the trade-offs present in OPLLs. The integrated POs discussed in this work could be an answer to the need for tuneable, portable, cost- and energy-efficient THz sources that can operate at room temperature. Photonic-enabled emitters have the potential to overcome the limitations of conventional emitters, thereby accelerating the development of coherent THz technology and its applications in spectroscopy, sensing, security and short-range broadband wireless communications.

Despite the extensive research within the field of resource management in monostatic multifunction radar, the resource management techniques for a multifunction radar network are still in their infancy. More specifically, a network which has the ability to switch modes between monostatic and bistatic configuration may potentially be able to capitalise on the advantages of both configurations. This is a gap which has been identified and is the aim of this thesis to explore. The research within this thesis begins by exploring the advantages provided by a bistatic configuration over a monostatic configuration. The conclusions from the initial research are carried forward to design a complete resource management framework for a multifunction radar network consisting of active and passive nodes. The resource management framework is broken into two sub-problems, resourceallocation and the scheduling problem. The resource-allocation problem deals with the task parameter selection methodology to optimally distribute the finite resources. This incorporates the concept of information sharing, which can be considered as a subset of information fusion theory, to delegate a given task to the best suited sensor within the network. A Quality of service framework is utilised to solve the resource-allocation problem where the resulting algorithm is referred to as APNQ-RAM. The scheduling problem is solved by deploying an earliest deadline first scheduler on master-slave architecture where the resulting algorithm is named as MS-EDFS. The research has also explored the impact of the networks geometry and the number of nodes on the network’s performance in tracking and surveillance functions. The research shows significant advantages in terms of tracking and surveillance performance provided by such a network in comparison to a monostatic configuration functioning on its own.

The work presented in thesis is dedicated primarily to improving the optical properties of vanadium dioxide (a thermochromic material used for energy-saving window applications) through the incorporation of bio-mimetic sub-wavelength nanostructures at the air–glass interface. The focal points of our work have been: firstly, to identify where in the world these types of smart windows are most effective; secondly, to synthesise vanadium dioxide and characterise its optical properties; thirdly, to develop design rules for nanostructured Motheye Smart Windows; fourthly and finally, to fabricate the desired motheye nanostructures in glass.

Recent advances in the field of neuroscience have suggested that new generation brain computer interfaces demand a critical step in biomedical signal processing requiring online/on-chip spike sorting. Spike sorting is the process of grouping signals from an individual neuron by grouping action potentials (spikes) into a specific cluster based on the similarity of their shapes. The extraction of single-unit activity by sensors at a distance from specific neurons is necessary for a wide range of clinical applications such as disorder treatments, muscular stimulation (e.g., epidural spinal cord stimulation for treatment acceleration), cochlear implant and neural prostheses. A brain machine interface, for example, can potentially substitute the missing motor pathway/sensory information between the motor cortex and an artificial limb. With the aim of developing an energy-efficient spike sorting chip for hardware implantable systems, this thesis introduces a new feature extraction method based on extrema analysis (positive and negative peaks) of spike shapes and their discrete derivatives. The proposed method runs in real-time and does not require any offline training. Compared to other methods it offers a better tradeoff between accuracy and computational complexity using online sorting. It additionally eliminates multiplications which are computationally expensive, power hungry and require appreciable silicon area. A minimum power limit for implantable neural front-end interfaces is also derived. It involved: 1) system level optimization - the front-end specifications including the bandwidth, data converter resolution and sampling rate were defined by exploring the effect of the parameters on spike sorting via a standard spike bank; 2) block level optimization - The front-end power was minimized by using an opamp-less cyclic converter; and 3) estimating the power limit equation of the frontend. The new optimization methodology addresses the future demands of neural recording interfaces. Finally the thesis presents the design, implementation and testing of the first generation of an adaptive spike sorting processor. It enhances the accuracy-power characteristics by employing self-calibration of processing features. The chip prototype was fabricated in a 180-nm CMOS technology. It achieves an overall clustering accuracy of 84.5% using a standard spike data bank and has a power consumption of 148-μW from 1.8-V supply voltage. The fabricated spike processor has almost 10%higher clustering accuracy than the state-of-the-art. Measurements show good power-performance characteristics compared to the state-of-the-art online and offline clustering methods.

Multistatic radar system provides a great number of advantages over monostatic radar systems via exploitation of geometrical diversity which gives the ability to view targets from multiple perspectives. These advantages include target signature enhancement, improved detection, higher immunity against countermeasures and provide more information about targets and hence enhance targets classification. Furthermore, the passive receive-only nodes are more compact, hence lighter weight, and can be mounted on UAV which leads to a great advantage to surveillance systems and military applications. Over the last two decades, University College London and the University of Cape Town have collaborated to develop multistatic radar systems. This collaboration resulted in developing an S-band system (NetRAD). Recently, a new version of the system (NeXtRAD) that operates in X and L bands has been developed. The NeXtRAD system has two channels at X-band which allows for receiving fully polarimetric data from clutter and targets utilising dual-polarised antenna. The first addressed task of this work was to investigate all possible antenna candidates to be used for the NeXtRAD system. Resonant SWGAA was chosen among the antenna options as it fit best all desired criteria and due to its relative design simplicity, high power-handling capabilities and cost-effective fabrication. The SWGAA can be designed to be a dual-polarised antenna. The procedures for designing a low sidelobe level (SLL) S-band SWGAA are demonstrated in this work. The azimuth beamwidth of a SWGAA is controlled by the number of slots carved in a waveguide. Eight slots distributed around the centre-line of waveguide broad wall found to meet the desired beamwidth. Four SWGAAs were designed and fabricated. The anechoic chamber measurements of each SWGAA showed excellent agreement with the simulation results. A single element SWGAA has a fan elevation beamwidth. This beamwidth has to be narrowed to achieve the desired width. Stacking identically designed SWGAAs was found to be an effective and simple method to narrow the antenna elevation beamwidth. The four SWGAAs were stacked on top of each other. The mitigation of mutual coupling between stacked SWGAAs was investigated. A cost-effective method of inserting dielectric sheets between stacked SWGAAs helped in mitigating the mutual coupling and assist in arriving at the desired antenna performance. The stacked SWGAAs shows a very good performance with very low SLL and high polarisation purity (low cross-polarisation level). The stacked SWGAAs antenna performance was validated in field experiments and compared to similar characteristics antenna. The SWGAA shows better performance compared to the other antenna. A new simple and efficient design of a dual-polarised SWGAA by having two similar set of stacked waveguides with one set rotated by 90ᵒ relative to the other one was proposed. Two designs with two different elevation beamwidths were simulated. Both designs showed excellent performance that met all the desired criteria. The same designs and tests procedures were followed in designing and testing the X-band SWGAA and simulating dual-polarised antennas. No field experiments were performed using this antenna as the NeXtRAD system is based at UCT and no access to any other X-band radar system at UCL.