Optical networks on-chip (ONoC) have been proposed to reduce power consumption and increase bandwidth density in high performance chip multiprocessors (CMP), compared to electrical NoCs. However, as buffering in an ONoC is not viable, the end-to-end message path needs to be acquired in advance during which the message is buffered at the network ingress. This waiting latency is therefore a combination of path setup latency and contention and forms a significant part of the total message latency. Many proposed ONoCs, such as Single Writer, Multiple Reader (SWMR), avoid path setup latency at the expense of increased optical components. In contrast, this thesis investigates a simple circuit-switched ONoC with lower component count where nodes need to request a channel before transmission. To hide the path setup latency, a coherence-based message predictor is proposed, to setup circuits before message arrival. Firstly, the effect of latency and bandwidth on application performance is thoroughly investigated using full-system simulations of shared memory CMPs. It is shown that the latency of an ideal NoC affects the CMP performance more than the NoC bandwidth. Increasing the number of wavelengths per channel decreases the serialisation latency and improves the performance of both ONoC types. With 2 or more wavelengths modulating at 25 Gbit=s , the ONoCs will outperform a conventional electrical mesh (maximal speedup of 20%). The SWMR ONoC outperforms the circuit-switched ONoC. Next coherence-based prediction techniques are proposed to reduce the waiting latency. The ideal coherence-based predictor reduces the waiting latency by 42%. A more streamlined predictor (smaller than a L1 cache) reduces the waiting latency by 31%. Without prediction, the message latency in the circuit-switched ONoC is 11% larger than in the SWMR ONoC. Applying the realistic predictor reverses this: the message latency in the SWMR ONoC is now 18% larger than the predictive circuitswitched ONoC.

Real-world image processing tasks often involve various image modalities captured by different sensors. However, given that different sensors exhibit different characteristics, such multi-modal images are typically acquired with different resolutions, different blurring kernels, or even noise levels. In view of the fact that images associated with the same scene share some attributes, such as edges, textures or other primitives, it is natural to ask whether one can improve standard image processing tasks by leveraging the availability of multimodal images. This thesis introduces a sparsity-based machine learning framework along with algorithms to address such multimodal image processing problems. In particular, the thesis introduces a new coupled dictionary learning framework that is able to capture complex relationships and disparities between different image types in a learned sparse-representation domain in lieu of the original image domain. The thesis then introduces representative applications of this framework in key multimodal image processing problems. First, the thesis considers multi-modal image super-resolution problems where one wishes to super-resolve a certain low-resolution image modality given the availability of another high-resolution image modality of the same scene. It develops both a coupled dictionary learning algorithm and a coupled super-resolution algorithm to address this task arising in [1,2]. Second, the thesis considers multi-modal image denoising problems where one wishes to denoise a certain noisy image modality given the availability of another less noisy image modality of the same scene. The thesis develops an online coupled dictionary learning algorithm and a coupled sparse denoising algorithm to address this task arising in [3,4]. Finally, the thesis considers emerging medical imaging applications where one wishes to perform multi-contrast MRI reconstruction, including guided reconstruction and joint reconstruction. We propose an iterative framework to implement coupled dictionary learning, coupled sparse denoising and k-space consistency to address this task arising in [5,6]. The proposed framework is capable of capturing complex dependencies, including both similarities and disparities among multi-modal data. This enables transferring appropriate guidance information to the target image without introducing noticeable texture-copying artifacts. Practical experiments on multi-modal images also demonstrate that the proposed framework contributes to significant performance improvement in various image processing tasks, such as multi-modal image super-resolution, denoising and multi-contrast MRI reconstruction.

Lateral Flow Immunoassays (LFIAs) are biosensors, which among others are used for the detection of infectious diseases. Due to their numerous advantages, they are particularly suitable for point of care testing, especially in developing countries where there is lack of medical healthcare centers and trained personnel. When the testing sample is positive, the LFIAs generate a color test line to indicate the presence of analyte. The intensity of the test line relates to the concentration of analyte. Even though the color test line can be visually observed for the accurate quantification of the results in LFIAs an external electronic reader is required. Existing readers are not fully optimized for point-of-care (POC) testing and therefore have significant limitations. This thesis presents the development of three readout systems that quantify the results of LFIAs. The first system was implemented as a proof of concept of the proposed method, which is based on the scanning approach without using any moving components or any extra optical accessories. Instead, the test line and the area around it, are scanned using an array of photodiodes (1 × 128). The small size of the pixels gives the system sufficient spatial resolution, to avoid errors due to positioning displacement of the strip. The system was tested with influenza A nucleoprotein and the results demonstrate its quantification capabilities. The second generation system is an optimized version of the proof of concept system. Optimization was performed in terms of matching the photodetectors wavelength with the maximum absorption wavelength of the gold nanoparticles presented in the tested LFIA. Ray trace simulations defined the optimum position of all the components in order to achieve uniform light distribution across the LFIA with the minimum number of light sources. An experimental model of the optical profile of the surface of LFIA was also generated for accurate simulations. Tests of the developed system with LFIAs showed its ability to quantify the results while having reduced power consumption and better limit of detection compared to the first system. Finally, a third generation system was realized which demonstrated the capability of having a miniaturized reader. The photodetector of the previous systems was replaced with a CMOS Image Sensor (CIS), specifically designed for this application. The pixel design was optimized for very low power consumption via biasing the transistors in subthreshold and by reusing the same amplifier for both photocurrent to voltage conversion and noise cancellation. With uniform light distribution at 525 nm and 76 frames/s the chip has 1.9 mVrms total output referred noise and a total power consumption of 21 μW. In tests with lateral flow immunoassay, this system detected concentrations of influenza A nucleoprotein from 0.5 ng/mL to 200 ng/mL.

A Transparent Cathode Discharge has been studied, with the aim of identifying operating regimes in which ions and neutrals are reactive species. The device consists of a cylindrical, powered cathode grid, that has generally been operated within a grounded anode grid, held concentric with the cathode. The principal mode structure evident across the operating pressure range is found to represent discharges sustained respectively by the activity of electrons at higher pressures, and ionic and neutral species at lower pressures. This is ascertained by analysis of the optical emission from the plasma, showing processes of excitation to be caused by these species. The discharge occurring at higher pressures, of tens to hundreds mTorr, has been characterised as an electron-driven 'cathode-confined' mode, that is shown to be sustained in part by the so-called 'hollow cathode effect'. The discharge occurring at lower pressures, of units to tens mTorr, is characterised as an 'ion beam' mode, composed of radially-convergent ion beams, from which Doppler-shifted light provides direct evidence for the presence of energetic ions and neutrals. Quasi-spherical space charge objects are observed to appear within the cathode, during operation at moderately low pressures of units to tens mTorr, and cathode currents of tens of mA. Larger space charge objects, termed 'fireballs', are observed to appear at a narrower range of conditions, and to show some analogous characteristics to these. Analysis of distributions of current and optical emission indicate the fireball objects to have a more complex internal structure. The physical dependences shown by the different operating modes upon the geometry of the electrode arrangement have been identified, so that the type and stability of plasma produced may be engineered by informed design. Discharge apparatus configurations have been identified that might be favourable for the creation of additionally-stable ion- and neutral-driven plasmas, and also for higher-power operation in convergent geometries.

This project studies the conversion of CO2 into fuels and chemicals in a dielectric barrier discharge (DBD) reactor. CO2, H2 and CH4 have been used as reactants, and special attention has been paid on understanding the plasma-catalytic synergy when a catalyst is placed in a plasma discharge. CO2 and CH4 are major greenhouse gases, responsible for the global greenhouse effect and climate change. The overall aim of this project is to initiate CO2 hydrogenation and biogas reforming at ambient temperature and atmospheric pressure by using plasma-catalysis. In this project, non-thermal plasma has been generated in a DBD reactor with and without a packed-bed of catalyst, enabling the CO2 conversion to be investigated under three conditions: Plasma alone, thermal catalysis and plasma-catalysis. Transitional metal catalysts such as Cu, Co, Mn, and Ni supported on Al2O3 and SiO2 have been screened, and their performance in the CO2 hydrogenation and biogas reforming have been compared under the three conditions. The synergy between non-thermal plasma and catalysts has been clearly identified. The effects of a catalyst’s properties and operational parameters on the reactions have also been studied. The project starts by the investigation of CO2 hydrogenation with H2. Results showed that reverse water-gas shift reaction and CO2 methanation were dominant in the plasma CO2 hydrogenation process. Compared to plasma CO2 hydrogenation without a catalyst, the combination of plasma with Cu/Al2O3, Mn/Al2O3 and Cu-Mn/Al2O3 catalysts enhanced the conversion of CO2 by 6.7% to 36%. The Mn/Al2O3 catalyst showed the best catalytic activity, as it increased the CO yield by 114% and the energy efficiency of CO production by 116%. The Ni/Al2O3 was even better than the Mn/Al2O3 catalyst, while its presence in the DBD reactor has clearly demonstrated a plasma-catalytic synergy at low temperatures. In addition, the introduction of argon in the reaction has enhanced the conversion of CO2, the yield of CO and CH4 and the energy efficiency of the plasma process. The formation of metastable argon (Ar*) in the plasma has created new reaction-routes which made a significant contribution to the enhanced CO2 conversion and CH4 yield. Biogas reforming has also been initiated at ambient temperatures by non-thermal plasma. The combination of plasma with the Co/Al2O3, Cu/Al2O3, Mn/Al2O3 and Ni/Al2O3 catalysts significantly enhanced CH4 conversion and showed a plasma-catalytic synergy for CH4 conversion and overall energy efficiency of the process. The best CH4 conversion of 19.6% and syngas production have been achieved over the Ni/Al2O3 catalyst at a discharge power of 7.5 W and a gas flow rate of 50 ml min-1. Moreover, the addition of K-promoter into the catalyst has further improved the performance of the Ni/Al2O3 catalyst. A conclusion of the findings of this project and outlook for further work is presented in Chapter seven, where it is concluded that non-thermal plasma has initiated the CO2 hydrogenation and biogas reforming at lower temperatures, comparing with thermal catalytic processes. The combination of plasma and catalyst has further improved the performance of the hydrogenation processes, in terms of conversion, yield, and energy efficiency, while significant synergy between DBD plasma and catalysts has been observed. By upgrading the catalyst and adjusting the operational parameters (e.g. molar ratio of feed gas, preparation method of catalyst, composition of catalyst, and promoters), the plasma-catalytic CO2 hydrogenation and biogas reforming processes can be further optimised.

No description available.

No description available.

No description available.

Considerable research efforts are being carried out worldwide to develop technologies which meet the increasing demand for the efficient utilisation of energy resources. Modern applications, such as renewable energy and electrical vehicles, place a premium on electro-mechanical energy conversion in a power dense and high efficiency manner. Magnetic gears (MG) and magnetically geared machines, offer an attractive alternative to existing systems which may favour the combination of a high speed electrical machine with a mechanical gearbox. This has led to the opportunity to use Pseudo Direct Drives (PDDs) and MGs to be developed for use on an industrial scale. Therefore, in this thesis techniques for facilitating the manufacture and robustness of PDDs are presented, for both radial and axial field topologies. This includes use of alternative windings and soft magnetic composites. PDDs and MGs has so far mainly been developed in the radial topology and little attention has been given to axial topologies. The pole piece (PP) rotor required for MG operation, represents the main difference between PDD/MG and a conventional electrical machine. As such the PP shape and supporting structures have been investigated both in terms of electromagnetic and mechanical performance. Furthermore, detailed electromagnetic and thermal design and analysis of an axial field PDD (AFPDD) with improved robustness was undertaken, and a prototype was manufactured to demonstrate the operation of the AFPDD and validate the predictions.

No description available.