The user growth in Internet services for the past years has caused a need to re-think methods for user authentication, authorization and accounting for network providers. To deal with this growing demand for Internet services, the underlying user authentication systems have to be able to, among other things, handle load spikes. This can be achieved by using loadbalancing, and there are both adaptive and non-adaptive methods of loadbalancing. This case study compares adaptive and non-adaptive loadbalancing for user authentication in terms of average throughput. To do this we set up a lab where we test two different load-balancing methods; a non-adaptive and a adaptive. The non-adaptive load balancing method is simple, only using a pool of servers to direct the load to in a round-robin way, whereas the adaptive load balancing method tries to direct the load using a calculation of the previous requests.

Bridge load rating is a procedure to determine the live load carrying capacity of a bridge. This rating is generally given out on a two-year period, which leaves the structural capacity unknown for this time interval. Conventional bridge load rating is obtained according to the bridge inspection results and commercial bridge rating software. However, this approach cannot effectively reflect actual live load carrying performance of the bridge, due to intrinsic limitation of visual inspection. Structural sensing has been utilized for measuring realistic structural behaviors to reflect the live load carrying capacity. However, this expensive and time-consuming process requires a known-weight vehicle and a substantial number of sensors under controlled full-scale field test conditions. In this research, a continuous live load performance index (LLPI) is proposed to monitor the live load capacity that the bridge can withstand without knowing the vehicle weight while also using a limited number of sensors. The LLPI uses existing bridge load rating methodology, in conjunction with experimental data and numerical simulations, to generate a value that describes the performance of the bridge due directly to the live load applied. Furthermore, the LLPI procedure utilizes an advanced state estimation algorithm, known as the Kalman Filter, to estimate the strain responses of the bridge at various locations while using a limited number of sensors. This procedure allows for an efficient structural health monitoring approach to determine the live load carrying capacity that the bridge can withstand. This research uses a lab-scaled truss structure with known properties for numerical and experimental validation. Because of this, this paper proposes a framework as to which the live load carrying performance can be monitored in real time. Future updates include testing on a real-life bridge structure while also determining optimal sensor placement for obtaining the LLPI. This research looks to develop a new live load performance index (LPPI) by considering: (1) the benefits and limitations of conventional bridge load rating approach, (2) the system identification and multi-metric data acquisition for the bridge structure, (3) numerical modeling and updating to best reflect the current dynamic properties of the bridge, (4) augmented Kalman Filter to estimate structural responses at various unknown locations, (5) LLPI formulation using experimental data, current bridge load rating methodology, and model-response estimations. The results obtained from this research provide a progressive live load capacity performance template to promote the advancement in civil infrastructure smart monitoring.

Autoliv has developed the Night Vision system, which is a safety system for use incars to improve the driver’s situational awareness during night conditions. It is areal-time system that is able to detect pedestrians in the traffic environment andissue warnings when there is a risk of collision. The timing behaviour of programsrunning on real-time systems is vital information when developing and optimisingboth hardware and software. As a part of further developing their Night Visionsystem, Autoliv wanted to examine detailed timing behaviour of a specific part ofthe Night Vision algorithm, namely the Tracking module, which tracks detectedpedestrians. Parallel to this, they also wanted a reliable method to obtain timingdata that would work for other parts of that system as well, or even other applications. A preliminary study was conducted in order to determine the most suitable methodof obtaining the timing data desired. This resulted in a measurement-based approachusing software profiling, in which the Tracking module was measured usingvarious input data. The measurements were performed on simulated hardwareusing both a cycle accurate simulator and measurement tools from the systemCPU manufacturer, as well as tools implemented specifically to handle input andoutput data. The measurements resulted in large amounts of data used to compile performancestatistics. Using different scenarios in the input data, we were able to obtain timingcharacteristics for several typical situations the system may encounter duringoperation. By manipulating the input data we were also able to observe generalbehaviour and achieve artificially high execution times, which serves as indicationson how the system responds to irregular and unexpected input data. The method used for collecting timing information was well suited for this particularproject. It provided the possibility to analyse behavior in a better waythan other, more theoretical, approaches would have. The method is also easilyadaptable to other parts of the Night Vision system, or other systems, with onlyminor adjustments to measurement environment and tools.

bjetivo. Evaluar la asociación entre síndrome metabólico (SM) y carga glicémica (CG) de la ingesta alimentaria en niños y adolescentes obesos atendidos en consulta endocrinológica de un hospital de referencia pediátrica en Lima, Perú. Población y método. Estudio transversal en niños y adolescentes obesos (índice de masa corporal ≥ percentil 95) de 10 a 15 años. Se clasificó como SM según los criterios de la International Diabetes Federation. Se aplicó un recordatorio de 24 horas y se calculó la CG total y por comida. Se evaluó la asociación entre los terciles de CG (comparación con el inferior) y SM usando razones de prevalencia ajustadas (RPa) por variables demográficas, antecedentes familiares, de actividad física y consumo total de carbohidratos en modelos de regresión de Poisson con varianza robusta. Resultados. De 273 niños y adolescentes obesos, 52,4% fueron varones y 94,9% fueron físicamente inactivos. La mediana de CG fue de 213 (164,8- 287,4) y la de ingesta calórica diaria fue 2275 (1851-3024) kcal, dada principalmente por carbohidratos (62%). La prevalencia de SM fue de 22,3%; los componentes con mayor prevalencia fueron la obesidad abdominal (81,7%) y los valores de lipoproteínas de alta densidad (HDL, del inglés high density lipoprotein) bajos (63,7%). Por último, se encontró una asociación entre el consumo elevado de CG y el riesgo de presentar SM (RPa 4,5; IC 95%: 1,3-15,3). Conclusiones. Existe una asociación entre el alto consumo de CG y la presencia de SM en niños y adolescentes con obesidad. / Objective. To study the association between the metabolic syndrome (MS) and the glycemic load (GL) of food intake among obese children and adolescents seen in consultation by the endocrinology team in a pediatric referral hospital in Lima, Peru. Population and Method. Cross-sectional study among obese children and adolescents (body mass index ≥ 95 percentile), 10-15 years old. The MS was classified according to criteria of the International Diabetes Federation. A 24 hour reminder was used, and the overall and per meal GL was calculated. The association between the GL tertiles (comparison with the lower tertile) and the MS was assessed using prevalence ratios adjusted by demographic outcome measures, family history, physical activity and total carbohydrate consumption in Poisson regression models with a robust variance. Results. Out of 273 obese children and adolescents, 52.4% were male and 94.9% did not engage in any physical activity. Glycemic load median value was 213 (164.8-287.4) and the daily calorie intake value was 2275 (1851-3024) kcal, consisting mainly of carbohydrates (62%). MS prevalence was 22.3%; the most prevalent components were abdominal obesity (81.7%) and low values of high density lipoprotein (HDL) (63.7%). Lastly, an association was observed between a high consumption of GL and the risk of developing MS (aRP 4.5; 95% CI: 1.3-15.3). Conclusions. There is an association between a high consumption of GL and the presence of MS among obese children and adolescents.

This research aimed to provide insight into the use and possible value of purpose built multimedia computer software for the study of English Literature. The software in question was developed in light of many years practical experience of teaching English Literature to secondary school students preparing for external examinations and was designed with the aim of improving their knowledge and understanding of particular works of literature. Informed by a critique of the main research findings about ICT use in learning and teaching since the period when computers were introduced into mainstream schools from the 1980s, the empirical research investigated two of the most prominent theoretical and practical perspectives that have been applied to understanding the relationship between educational resources and learning: Learning Styles Theory and Cognitive Load Theory. These two approaches and their associated instrumentation were applied in a quasi-experimental controlled empirical study in four schools in the north-east of England where the multimedia software was used with groups of students embarking on a study of Shakespeare's Macbeth for GCSE examination. Learning Styles theory and the instrumentation used (Kolb's LSI and Honey & Mumford's LSQ) proved less successful than Cognitive Load Theory in demonstrating reliability and validity and therefore in explaining the relationship between different instructional resources and individual learning. The theoretical integrity and usefulness of these two approaches is discussed and, in particular, the rationale behind the continued use of Learning Styles was explored via interview with school faculty who gave reasons of face validity; the pressure from external inspection; the mechanisms through which they were held professionally accountable; senior management and institutional policy; the legacy of initial teacher training; and established classroom practice. Students using the multimedia software demonstrated improvements in their knowledge and understanding of Macbeth equivalent on average to one GCSE grade above those not using the multimedia resource and Cognitive Load Theory was found to be successful in explaining this and in predicting the relationship between instructional resource and the learning gains of individuals. Limitations of the study are drawn along with conclusions for further research and for enhancing teaching and learning with multimedia resources.

Cette recherche se concentre sur l'optimisation d'un micro-réseau en interaction avec le réseau électrique intelligent. Il s'agit de la recherche de solutions optimales pour la conception d'un micro-réseau afin de minimiser les coûts, d'une part, et la possibilité augmenter 1'utilisation des sources renouvelables, d'autre part. La supervision, doit traiter la prise en compte des incertitudes dans la gestion prédictive optimisée des flux de puissance. / This thesis study focuses on a DC microgrid building-integrated satisfying the power balance at the local level and supplying DC loads during both, grid-connected and isolated operation modes. Considering that energy management can be defined as a group of different control strategies and operational practices that together with the new physical equipment and software solutions aims to accomplish the objectives of energy management, the main objective of this thesis is to define the energy management strategies for the building-integrated DC microgrid, aiming to keep the bus voltage stable as well as to reduce the energy cost to the end users and the negative impact to the main grid. Therefore, this research work focuses to optimize and develop the implementation of the designed controller of building-integrated DC microgrid. The proposed DC microgrid consists of PV building-integrated sources, a storage system, a main grid connection for the grid-connected mode and a micro turbine for the off-grid or isolated mode, and a DC load (electric appliances of a tertiary building). The bidirectional connections with the main grid and the storage aim to supply the building’s DC appliances, and sell or store the energy surplus. The results validate the operation of the whole system, ensuring the capability of the proposed supervisory control to manage the energy power flow while ensuring voltage stability. Other goals concern the analyze of the proposed separation between optimization and real time power balance and the usage of the proposed load shedding/restoration algorithm in the microgrid environment are also validate. Regarding the technical contributions, the work of this thesis allowed the creation and the practical development of a test bench for microgrid based on PV sources emulator, which allows the repeatability conditions (closeness of the agreement between the results of successive measurements of the same solar irradiance and air temperature carried out under the same conditions of measurement) and reproducibility (closeness of the agreement between the results of measurements of the same solar irradiation and air temperature carried out under changed conditions of measurement). Numerous experimental tests were carried out and allowed the validation of the proposed concepts.

Abstract The natriuretic peptides are a family of hormones secreted by the heart. They play a fundamental role in salt and water balance and blood pressure regulation. Atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP) and C-type natriuretic peptide (CNP) are the known members of the mammalian natriuretic peptide family. A major stimulus for the secretion of cardiac natriuretic peptides is myocyte stretch. Therefore, the secretion of natriuretic peptides is increased in response to elevated blood volume. Natriuretic peptide production and release is also affected by several other factors, such as endothelin-1 (ET-1), acting in paracrine fashion. The aim of this study was to elucidate factors regulating the novel cardiac peptide hormone, salmon cardiac peptide (sCP), belonging to the family of natriuretic peptides. The role of mechanical load, temperature and ET-1 in sCP secretion and production was studied using in vitro (isolated perfused ventricle preparation) and in vivo methods. Comparisons between the natriuretic peptide systems in fish and mammals were done to clarify functional evolution of this hormone family. Salmon (Salmo salar) was selected as a model, since it has an outstanding adaptability to wide variations in environmental salinity and has developed defence mechanisms against volume or salt load. The results showed that salmon ventricle stores large amounts of the prohormone of sCP, whereas the secreted form is the mature 29-amino acid form. The N-terminal fragment of pro-sCP is co-secreted with sCP in equimolar amounts. sCP is released rapidly in response to appropriate stimulus, whereas induction of its gene expression is slower. Mechanical load is an important regulator of sCP secretion. Temperature also plays a major role in regulating sCP plasma concentration by affecting its elimination from circulation. Additionally, ET-1 is a potent secretagogue of the sCP system and an inotropic agent in salmon heart. Furthermore, the present results reveal remarkable synergism between the cardiac effects of ET-1 and β-adrenergic stimulation. In conclusion, the sCP system in salmon ventricle largely resembles the ANP system in mammalian atrium, while also having specific characteristics, such as a regulated ventricular natriuretic peptide secretion pathway. Therefore, the sCP system offers a unique model for studying mechanisms of natriuretic peptide biology.

The rate that HIV-infected individuals progress to AIDS and death varies greatly. Viral load taken during the asymptomatic phase of the disease is one of the best-known predictors of HIV progression rate and transmission risk, and is known to be in uenced by both host and environmental factors. However, the role that the virus itself plays in determining the viral load is less clear. Previous studies have attempted to quantify the amount the viral genome in uences viral load, or the heritability of viral load, using transmission pairs and phylogenetic signal in small sample sizes, but have produced highly disparate estimates. E cient and accurate methods to estimate heritability have been utilised by quantitative geneticists for years, but are rarely applied to non-pedigree data. Here, I present a novel application of a population-scale method based in quantitative genetics to estimate the heritability of viral load in HIV using a viral phylogeny. This new phylogenetic method allows the inclusion of more samples than ever previously used, and avoids confounding e ects associated with transmission pair studies. This new method was applied to the two largest HIV subtypes found in the UK, subtypes B and C, using sequences and clinical data from UK-wide HIV databases. For subtype B (n=8,483) and C (n=1,821), I estimated that 5.7% (CI 2.8{8.6%) and 29.7% (CI 14.8{44.7%) of the variance in viral load is determined by the viral genome, respectively. These estimates suggest that viral in uence on viral load varies greatly between subtypes, with subtype C having much larger viral control over viral load than subtype B. I expanded the phylogenetic method to test whether the component of the viral load determined by the virus has changed over time. In subtype B, I foundevidence of a small but signi cant decrease in the viral component of viral load of -0.05 log10 copies/mL/yr. I built a stochastic, individual-based model capable of simulating a realistic HIV epidemic, with heritable viral loads that in uence transmission and disease progression, capable of generating data sets to assess the accuracy of phylogenetic methods. This was successfully used to generate epidemics approximating those in a small African village and a Western `men who have sex with men' community under a variety of conditions. To test the accuracy of the new phylogenetic heritability estimation method, simulated datasets were generated with the heritability of viral load set at values of 30%, 50%, 70%, and 90%. Unfortunately, complications in the heritability equation used prevented full assessment of the new phylogenetic method on the simulated data. Future development of the model will enable simulation of realistic viral loads under varying heritability values, enabling simulation of data sets that can be used to test this and other heritability estimation methods. This new phylogenetic method allows accurate estimation of heritability in large datasets, and has provided valuable insight into the viral in uence on viral load in HIV.

This study expanded the current literature by assessing PTSD in relation to reactivity and recovery from negative emotional arousal among OEF/OIF/OND Veterans. Cardiac impedance was employed during a speech task and a trauma imagery procedure. Those in the PTSD-S group displayed lower SBP and higher TPR reactivity relative to the PTSD- and PTSD+ groups; lower CO reactivity relative to the PTSD+ group; and more CO recovery than those in the PTSD+ group to the trauma task. For speech task, Veterans in the PTSD-S group exhibited lower HR reactivity for both speech preparation and delivery than those in the PTSD- group. Depression was not a significant mediator in the relation between PTSD and reactivity. However, further analyses revealed that it served as a moderator between PTSD and reactivity during speech preparation (SBP, HR, and PEP reactivity), and speech delivery (HR, PEP, and CO reactivity). Simple slopes analysis revealed that depression was positively associated with HR and SBP (speech preparation) and HR (speech delivery) for those in the PTSD-S group. For those in the PTSD- group, depression was positively associated with PEP during the speech task (to include preparation) and negatively associated with SBP and CO (speech preparation) and HR and CO (speech delivery). For those in the PTSD+ group, depression was negatively associated with CO and positively associated with PEP. For the most part, Veterans in the PTSD-S group exhibited lower reactivity to both tasks than Veterans in the PTSD+ group or combat-exposed controls without PTSD. In light of the emerging evidence relating blunted reactivity to unhealthy behaviors and negative health outcomes (e.g., depression, obesity), it would appear that both extremes, exaggerated and diminished reactivity are maladaptive responses to stress and that the most optimally response to stress is a moderate reaction.

The Unified Modeling Language (UML) has quickly become the industry standard for object-oriented software development. It is being widely used in organizations and institutions around the world. However, UML is often found to be too complex for novice systems analysts. Although prior research has identified difficulties novice analysts encounter in learning UML, no viable solution has been proposed to address these difficulties. Sequence-diagram modeling, in particular, has largely been overlooked. The sequence diagram models the behavioral aspects of an object-oriented software system in terms of interactions among its building blocks, i.e. objects and classes. It is one of the most commonly-used UML diagrams in practice. However, there has been little research on sequence-diagram modeling. The current literature scarcely provides effective guidelines for developing a sequence diagram. Such guidelines will be greatly beneficial to novice analysts who, unlike experienced systems analysts, do not possess relevant prior experience to easily learn how to develop a sequence diagram. There is the need for an effective sequence-diagram modeling technique for novices. This dissertation reports a research study that identified novice difficulties in modeling a sequence diagram and proposed a technique called CHOP (CHunking, Ordering, Patterning), which was designed to reduce the cognitive load by addressing the cognitive complexity of sequence-diagram modeling. The CHOP technique was evaluated in a controlled experiment against a technique recommended in a well-known textbook, which was found to be representative of approaches provided in many textbooks as well as practitioner literatures. The results indicated that novice analysts were able to perform better using the CHOP technique. This outcome seems have been enabled by pattern-based heuristics provided by the technique. Meanwhile, novice analysts rated the CHOP technique more useful although not significantly easier to use than the control technique. The study established that the CHOP technique is an effective sequence-diagram modeling technique for novice analysts.