21684 research outputs found
Sort by
Fuzzing techniques for automated vulnerability detection in IoT firmware
Full text linkA security flaw in the firmware of microcontrollers (MCUs) can lead to devastating consequences. Finding and fixing these bugs before deployment is essential because patching them in the field is often difficult, expensive, or impossible. However, standard software testing techniques like fuzzing struggle with embedded firmware due to its tight coupling with specialized hardware, which makes testing slow, inaccurate, and inefficient. This thesis studies two key design choices: where tests run (emulation, Hardware-in-the-Loop (HIL), or on-device) and what feedback and inputs they use (control flow vs. data flow; generic vs. domain-specific). It moves testing from slow emulation to real hardware and replaces simple code coverage with data-flow guidance to drive bug finding. It also measures how new hardware features can prevent whole classes of bugs.
The approach is demonstrated through four linked contributions. First, Sizzler solves the input wasted problem by generating valid, domain-aware tests for Programmable Logic Controllers (PLCs) by deep learning model, so fuzzing effort is not wasted. Second, FuzzRDUCC improves feedback by tracking def-use chains, revealing subtle bugs that edge-based coverage can miss. Third, Hardfuzz brings this data-flow guidance onto real hardware, using hardware breakpoints for fast, consistent testing. Finally, a differential testing framework for MicroPython compares builds with and without architectural memory-safety features from CHERI and shows which bug classes they block.
These results show that firmware testing benefits from hardware-centric, data-flow-guided methods. These approaches yield smarter, domain-aware inputs; feedback that is more informative than edge coverage; and fast, consistent testing on real devices. It also provides clear evidence that architectural memory safety-exemplified by CHERI-can block whole classes of vulnerabilities. In short, the thesis shifts the goal from only finding bugs to also preventing them by design
Feeling confident: a history of self-feeling in twentieth-century England
Full text linkThis thesis explores the history of self-feeling in twentieth-century England. It argues that self-referential emotions such as self-esteem, self-confidence, and inferiority-feeling underpinned new conceptualisations of childhood in this period. Ideas about how individuals feel – or ought to feel – about themselves influenced new understandings of children, positioned adults as emotional guardians, and spurred interventions designed to manage the self-concept of young people in England. This research challenges existing scholarship of self-feeling that positions self-esteem as a distinctively late twentieth-century American phenomenon linked to neoliberalism. It demonstrates that concern with self-feeling emerged much earlier in English culture as a contested mediator between individual development and social cohesion. Through examination of childrearing advice, education policy, grassroots initiatives, and child-led activism from the interwar period to the late 1980s, the thesis reveals how self-feeling operated not as a straightforward tool of individualisation but as a persistent and multifaceted source of anxious debate. Cultural concern with self-feeling reflected broader tensions about emotional vulnerability, educational purpose, and national progress. This thesis traces some of the processes through which psychological expertise became common sense, focusing on childhood as the primary site on which self-feeling was understood, negotiated, and applied. While adults were tasked with managing the self-feelings of children, children themselves emerged as complex negotiators who actively participated in constructing and transforming the emotional frameworks designed to regulate their development. By shedding light on these dynamics, this research offers a new lens for interpreting childhood emotion in modern England
A historical geography of mills in Scotland: innovation, adaptation, and inertia during Scotland’s Industrial Revolution
Full text linkMills have been an important feature of the Scottish landscape for centuries, beginning with the simple rural grain mill and developing into the iconic looming factories of the Industrial Revolution. The objective of this thesis was to chart the historical geography of mills in Scotland, with a primary focus on water-powered mills during the period of industrialisation from the late-eighteenth century through to the early-twentieth century. To achieve this objective the thesis has three overarching aims. First, to document mill distributions at a national scale. Second, to examine the geography of mills through their physical, social-political, and economic contexts, questioning the development paths of economic activity within given localities. Third, to consider the ways that locational decisions were governed by processes of inertia.
The approach taken was to reconcile macro- and micro-scale research, to map the various types of mill found in Scotland, followed by case studies from the cotton spinning industry, focussing on the Rothesay Cotton Mills, New Lanark Mills, Deanston Mills, Catrine Mills, Ballindalloch Mill, and Spinningdale Mill.
This thesis quantifies and presents the geographical distribution of mills in Scotland based on the First and Second edition Ordnance Survey maps and Historic Environment Scotland’s Canmore database. It then argues that natural endowments of the landscape shaped mill geographies and that water retained importance throughout the Industrial Revolution. It also argues that mill owners were making locational choices grounded in innovative economically and empirically-based decisions that, through adaptations to mill infrastructure and business practices, manifested in a dynamic form of inertia acting upon mill geographies. This thesis contributes to scholarship on industrialisation in Scotland, working with historic maps, and furthers our knowledge of the number, location, and history of industrial heritage assets
Understanding climate change adaptation strategies in social-ecological systems: the case of Bangladesh
No full textAbstract not currently available
Non-local resonant metasurface enhanced spectral filtering
Full text linkA metasurface is a quasi-two-dimensional micro-nano structure with subwavelength scale thickness. Owing to their compactness and the ability to integrate multiple functions, metasurfaces are seen as an excellent platform for optical field manipulation and photonrelated experiments. In this thesis the focus of the research is the emerging concept of nonlocal resonant metasurfaces. Appearing in array form, they demonstrate narrowband wavelength-selective control of the optical field through collective non-local resonant modes. Two important parameters are spectral and polarisation control, and these are the main topics of study in the thesis. The ability to manipulate the spectral and polarisation selective properties of metasurfaces has applications in sensing and several novel devices in the visible and infrared wavelength range are proposed and demonstrated.
In the visible spectral range, a novel plasmonic metasurface filter with a periodic dimericaperture hole structure on a metal film is proposed. It integrates colour-dependent filtering and polarisation-dependent transmission, enabling two functions simultaneously. Simulations show that the structure has improved transmittance compared to a conventional monomeric-aperture structure, exceeding 30% in the blue, green, and red channels, and achieves an orthogonal polarisation extinction ratio greater than 1000. Fabricated devices made by electron-beam lithography and dry etching exhibited approximately 30% transmittance in red, green and blue channels, a filtering linewidth of about 100 nm, and an average polarisation extinction ratio exceeding 20. A polarisation–colour palette was made to observe simultaneous modulation of colour and polarisation, and a colour-gamut diagram was plotted to show vivid colours with more than 76% sRGB coverage. The simultaneous colour–polarisation modulation was further applied to polarisation-dependent encrypted displays and the reproduction of micro-structured artwork.
In the infrared spectral range, a novel narrowband filter based on quasi–bound states in the continuum (q-BIC) is proposed. The device is composed of periodic nano-disks with offcentre holes along orthogonal axes to realise dual symmetry-breaking perturbations. The device therefore enables resonance-related linear-to-linear cross-polarisation coupling. Unidirectional transmission enhancement is achieved by tuning the coupling between eigenmodes through adjusting the size of the holes, showing a transmission exceeding the theoretical limit value of 25% derived from temporal coupled mode theory for single-layer, single-excitation metasurface. By extracting the cross-polarisation transmission channel, narrowband filtering is realised in transmission mode. Multipolar decomposition and mode- coupling related simulations explained the operating mechanism of the device. Devices fabricated using electron-beam lithography and dry-etching processes showed 30 % crosspolarisation conversion efficiency in Fourier-transform infrared microscope measurements. The experiments also exhibited a resonance linewidth of less than 60 nm at centre wavelength around 3500nm, corresponding to a quality-factor (Q) of approximately 60, which is an order of magnitude higher than previous experimental reports of free-space infrared filters based on plasmonic structures. The fabricated device was further applied to the reconstruction of infrared gas spectra. Experiments demonstrated real-time, in situ, spectral reconstruction of the semi-quantitative infrared absorption spectrum of butane gas, showing good agreement with the NIST ground-truth dataset.
A novel flat-band q-BIC metasurface working in the infrared spectral range is also proposed. High Q resonances are often accompanied by high sensitivity to the incident angle that is a limitation of non-local metasurfaces. In this study, band-structure engineering has been demonstrated through simulations. Using Brillouin-zone folding and symmetry breaking, strong coupling can be achieved between two modes that are not previously coupled because of a symmetry mismatch. The anti-crossing of the bands, because of the strong-coupling effect, adjusts the line shape of the band on which the mode lies in momentum space, thereby realising band-structure control hence exhibits a flat-band characteristic within an incidentangle range of approximately 15°. This new result presents opportunities for future experimental studies
Environmental controls on parasite infection dynamics in a multihost community at the wildlife-livestock interface
Full text linkWildlife-livestock interfaces are increasing globally, mainly due to the expansion of livestock production within a shrinking footprint for wildlife. Compressed ecosystems increase interactions of livestock, wildlife and humans, as well as the pathogens and parasites they carry. These interactions are core processes contributing to the emergence and spread of many infectious diseases and advancing our understanding of the direct and indirect ways that ecological interactions and environmental conditions shape parasite infection risk is a major priority under global change. Yet in many cases, we do not even know the composition of parasites that occur in different host species of livestock and wildlife and their transmission pathways occurring at interfaces, particularly from tropical settings. Migratory wildlife have large ecological effects along their migration routes, and with them may transfer parasites into new areas, spreading infectious diseases across management or political borders. This thesis synthesizes the role of migratory herbivores, vegetation and environmental factors such as rainfall, temperature and humidity in changing gastrointestinal nematodes infection risks across the landscape at the wildlife-livestock interfaces.
While migratory hosts move parasites and pathogens (transport effects) between areas, potentially infecting other host species, they can also have large effects on the environment via trophic effects in ways that may indirectly impact risk of transmission. Currently, we have little understanding of the magnitude, direction and net effects of migratory host’s transport and trophic effects in changing infection risks across ecosystems. In chapter 2, I used an experimental approach, and integrated framework that linked migration intensity and duration versus transport and trophic effects to study how migratory herbivores change pasture infection risks across the landscape. Migration intensity (the density of migrants in an area) and duration (the period that migrants remain in an area) determined the relative size and direction of transport and trophic effects on environmental parasite abundance in pasture. High migration intensity led to both strong transport and trophic effects upon environmental parasites. Grazing led to parasite consumption (hoovering effects) and changes to microclimate conditions, that resulted in a net decrease in availability of free-living stage of environmental parasites. Longer migration duration in an area increased both transport and trophic effects due to multiple dung depositions and multiple bouts of grazing. However, the trophic effects outweighed transport effects, and the resulting environmental parasite abundance was lower under high duration conditions compared to low duration conditions. The results suggest that grazing strongly modifies the infection risk for the subsequent hosts, and that both the intensity and duration of grazing, relative to the timing of parasite emergence, determine prevailing infection risk in pasture following herbivore migration.
Trees form an additional component of structural heterogeneity in savannas that also shape herbivore distribution and abundance, but we have a poor understanding of how heterogeneity induced by trees and herbivores, or their interactions, may drive infection risks across landscapes. In chapter 3, using a combination of observational and experimental approaches, I investigated how heterogeneity of grass cover induced by the herbivores’ trophic effects or by the presence and absence of tree cover, shaped infection risk of environmental parasites across the landscape. I found that shade availability from grass and tree canopy affected the abundance of environmental parasites by modifying microclimate conditions. The presence of shade from vegetation, such as from trees canopies and grasses, reduced the amount of sunlight reaching the soil surface grass layer by half or more, compared to areas without any vegetation cover. Furthermore, maximum temperature values were, on average, up to 8.5 °C higher in unshaded areas compared to shaded areas. As a result, abundance of environmental parasites was twice as high in shaded areas compared to unshaded areas. The observed microclimate effects on gastrointestinal nematode abundance also changed with aridity conditions, where overall there was an increase in environmental parasite abundance during wet conditions compared to dry conditions in all treatments, and the infection risk increased in unshaded shorter grass compared to unshaded long grass during wet conditions compared to dry conditions. My results suggest that, by changing vegetation structure, herbivore grazing reduces infection risk across the landscape by exposing free living stages of parasites to harsh weather conditions, resulting in lower abundance of environmental parasites. Trees, in contrast, by blocking sunlight radiation, increase infection risk across the landscape by lowering temperature and increasing humidity of grasses underneath, conditions that enhance survival of free-living stages, and lead to greater parasite abundance in the environment.
Overlap between wildlife and livestock is a common phenomenon that is currently increasing worldwide, but our understanding of ways in which overlapping wildlife and livestock are beneficial to each other under different contexts is limited. In chapter 4, I examined the consequences, in terms of infection, for livestock that overlapped spatially with migratory wildlife. Specifically, I used a natural experiment to evaluate how the intensity and composition of gastrointestinal nematode (GIN) infection in ruminant livestock changed depending on exposure to migratory wildebeest. I used a before-and-after impact design to sample GIN from livestock relative to the wildebeest migration. Wildebeest migration was associated with changes of livestock infection depending on the level of exposure to wildebeest and environmental conditions of the region where livestock occur. Livestock exposed to wildebeest migration had lower infection intensity compared to livestock not exposed to wildebeest migration. Goats’ infection intensity was higher in the wetter region (West) compared to drier region (East), while for cattle the infection was higher in dry region (East) compared to wetter region (West). Wildebeest migration also decreased species richness of worms in goats in areas grazed by wildebeest compared to areas not grazed by wildebeest, but post wildebeest migration also increased species richness in area exposed to wildebeest compared to pre-wildebeest migration. In contrast for sheep, there was an increase in worm species richness post wildebeest migration compared to pre wildebeest migration timepoint. The study findings suggest overlap between wildlife and livestock may be beneficial to livestock by reducing infection intensity or harmful by introducing new worms or increasing chances of rare worm species to infect livestock in different areas. Furthermore, environmental conditions, type of livestock, and livestock husbandry modified resulting infection intensity from wildlife and livestock overlap.
The synthesized knowledge about the interconnected roles of wildlife migration, vegetation structure and environmental conditions on parasite dynamics broadens our understanding of infectious disease epidemiology at the wildlife-livestock interface and contributes to pathways that promote wildlife-livestock co-existence, by quantifying impacts and benefits of wildlife and livestock co- existence. Furthermore, the knowledge could be used to improve intensive livestock production in tropical areas by identifying the conditions under which promoting co-grazing between livestock and migratory wildlife might reduce infectious disease in livestock and how different vegetation structures shape infection risks during wet and dry conditions
GasSight: a portable gas imaging camera
Full text linkMost gases do not absorb visible light, meaning they are invisible to the naked eye. Methane is one such invisible gas. According to NASA1 methane is the second most impactful greenhouse gas, but its short lifespan in the atmosphere means that reducing atmospheric methane could bring climate benefit on a geologically rapid timescale. Whilst natural sources of methane gas are common, much of the atmospheric methane is from anthropogenic sources, for example via its use in natural gas fuel. Leaks of methane along pipelines often occur, but are challenging to detect. Many optical techniques exist for the imaging and hence localisation of gas source, but each have various strengths and weaknesses in terms of detection sensitivity, detection range, ease of use and frame rate. In this thesis a video-rate gas imaging system is introduced that features a compact size, an infrared frame rate of around 15 fps, and a moderate detection range of up to 16 m. The system makes use of flood illumination at 1653 nm to probe an absorption line of methane, allowing a full scene to be investigated whilst maintaining eye-safety. The device was designed to be compact with a low mass of 3.15 kg, making it suitable as a drone payload for aerial pipeline surveillance. The research described throughout this thesis covers the work developing this prototype system based upon an earlier concept demonstrator, including the improvements made to performance and the successful drone mounted field trials. An improved further iteration of the system was designed to overcome the shortcomings of the prototype system, specifically in terms of sunlight rejection. The main benefit of this iterated system are its increased infrared frame rate of 100 fps, while maintaining comparable detection range and identical form factor. This thesis presents a gas imaging device that satisfies the need of close range detection, with portability and high frame rate. This device could work in tandem with lower frame rate systems capable of detection ranges exceeding 100 m in order to increase methane monitoring deployment worldwide and detect and reduce emissions
Segmentation variability and perfusion differences in contrast-enhancing and T2 defined glioblastoma regions using arterial spin labelling MRI perfusion
No full textAbstract not currently available
Collecting eighteenth-century dress, textiles, and the creation of early America: gender and curation in twentieth-century Colonial Revival museums
No full textAbstract not currently available
On quantum graph theory: non-commutative graph theory
Full text linkThis research paper aims to introduce quantum graphs to newcomers. We adopt a clear and concise linear algebraic approach to simplify the concepts of quantum graphs. Our main perspective is viewing a quantum graph as a quantum adjacency matrix operator on a finite-dimensional C*-algebra (which can simply be thought of as a direct sum of matrix algebras), with the inner product defined by choosing a faithful positive linear functional on each matrix algebra summand of the direct sum. Our main result presents practical formulae for identifying isomorphic single-edged quantum graphs. An immediate corollary to this is that there is an infinite number of isomorphisms for a single-edged quantum graph on Mₙ(C) for n > 2, which is surprising since all single-edged quantum graphs are isomorphic to one another for n = 2
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇