Distributed real-time database systems demand consistency and timeliness. One approach for this problem is eventual consistency which guarantees local consistency within predictable time. Global consistency can be reached by best effort mechanisms but for some scenarios, e.g. an alarm signal, this may not be suffcient. Bounded delay replication, which provides global consistency in bounded time, ensures that after the local commit of a transaction updates are propagated to and integrated at any remote node within bounded time. The DRTS group at the University of Skövde is working on a project called DeeDS, which is a distributed real-time database prototype. In this prototype, eventual consistency with as soon as possible (ASAP) replication is implemented. The goal of this dissertation is to further develop replication in this prototype in coexistence to the existing eventual consistency which implies the extension of both the theory and the implementation. The main issue with bounded time replication is to make all parts, which are involved in the replication process predictable and simultaneously support eventual consistency with as soon as possible replication.

Doctor of Philosophy / Department of Statistics / Weixing Song / In this dissertation, goodness-of-fit tests are proposed for checking the adequacy of parametric distributional forms of the regression error density functions and the error-prone predictor density function in measurement error models, when replications of the surrogates of the latent variables are available. In the first project, we propose goodness-of-fit tests on the density function of the regression error in the errors-in-variables model. Instead of assuming that the distribution of the measurement error is known as is done in most relevant literature, we assume that replications of the surrogates of the latent variables are available. The test statistic is based upon a weighted integrated squared distance between a nonparametric estimate and a semi-parametric estimate of the density functions of certain residuals. Under the null hypothesis, the test statistic is shown to be asymptotically normal. Consistency and local power results of the proposed test under fixed alternatives and local alternatives are also established. Finite sample performance of the proposed test is evaluated via simulation studies. A real data example is also included to demonstrate the application of the proposed test. In the second project, we propose a class of goodness-of-fit tests for checking the parametric distributional forms of the error-prone random variables in the classic additive measurement error models. We also assume that replications of the surrogates of the error-prone variables are available. The test statistic is based upon a weighted integrated squared distance between a non-parametric estimator and a semi-parametric estimator of the density functions of the averaged surrogate data. Under the null hypothesis, the minimum distance estimator of the distribution parameters and the test statistics are shown to be asymptotically normal. Consistency and local power of the proposed tests under fixed alternatives and local alternatives are also established. Finite sample performance of the proposed tests is evaluated via simulation studies.

The purpose of this study was to examine the consistency of near-infrared spectroscopy (NIRS) measures associated with a 10 minute vascular occlusion of the thigh during 5 consecutive days. A secondary purpose was to examine if the repeated daily occlusions modify the vascular response to such occlusions. The NIRS chromophores of oxygenated hemoglobin (HbO2), deoxygenated hemoglobin (HHb), total hemoglobin (tHb), hemoglobin difference (HbDiff), and total saturation index (TSI) were monitored at the vastus lateralis muscle during each testing session. Participants were female (age 18-30 yrs) and were randomly assigned to one of two groups. The Test Group (age 23 ± 2yrs) completed an occlusion test every day for 5 consecutive days while the Control Group (age 21 ±1.6yrs) was tested on days 1, 3 and 5. Each testing session involved 10 minutes rest in a supine position, 10 minutes of femoral artery occlusion (preset at 250mmHg), 10 minutes of recovery. At onset of occlusion, the rapid cuff inflation system inflated in 0.3 seconds to the preset 250mmHg. Following the 10 minutes of occlusion, the cuff was deflated in 0.3 seconds. Heart rate was monitored throughout all testing sessions and mean arterial pressure was calculated and compared between groups. There were no significant group or day main effects found, nor any significant interactions for the following hemoglobin chromophores measures: HbO2, HHb, HbDiff, and TSI. In all participants, the NIRS response pattern to rest, occlusion and recovery was consistent for all testing days regardless of group. The response pattern to the occlusion initiated by the rapid cuff inflation was as follows: HbO2, tHb, HbDiff, and TSI decreased then slowly plateaued over the 10 minutes while HHb increased initially and then plateaued. Following the rapid cuff deflation, HbO2, tHb, HbDiff and TSI increased to their maximum reperfusion levels and then plateaued over the remainder of the 10 minutes of recovery. During that same period, HHb decreased initially upon reperfusion and then plateaued. Unlike the other measures, TSI and tHb demonstrated inconsistencies in their response pattern regardless of group or day. The consistency of hemoglobin chromophores response pattern to daily 10 minute vascular leg occlusions over 5 consecutive days is an important finding of this study. This response indicates that there is no apparent change in vascular response to repeated occlusions and has implications for exercise studies that use occlusion in their methodology. The study also emphasizes the inconsistency of tHb and TSI as primary hemoglobin response measures to occlusion. Similarly, the results demonstrate the potential misinterpretation of results if only relying on one hemoglobin measure during occlusion. / Graduate

Across avian systems, plumage ornamentation is often considered a signal of individual fitness, condition, sex, or status, and varies due to genetics or environmental sources. In species with structural coloration, plumage variation results from differences in the amount of energy allocated to feather growth during molt, presenting a unique opportunity to study the link between individual quality and ornamentation. In cooperative breeding species, such as the Florida Scrub-Jay (Aphelocoma coerulescens), the role of structural color as a signal is particularly important because helpers may delay breeding for one to several years and competition for space is high. Florida Scrub-Jay juveniles are sexually dimorphic in the ultraviolet range, and plumage color predicts social dominance and condition, but not adult reproductive fitness. Little is known about the consistency of ultraviolet reflectance across molts, and I tested the following questions: 1) does plumage color change across molts within individuals; 2) how do environmental variables and stress affect structural color; and 3) does adult color or change in color predict reproductive fitness? I measured relative change in color within individuals by comparing their juvenile and first set of adult feathers after pre-basic molt. I used several measures of nutritional condition, social dynamics, habitat quality, and parasite infection to model color, and I experimentally administered corticosterone during pre-basic molt to examine the effects of increased stress on color. Plumage reflectance was compared with breeding status to investigate the role of color in breeding space acquisition. Plumage reflectance was significantly different within individuals across molts, but juvenile and adult color were correlated. Adults were significantly less bright than juveniles, with higher proportions of UV chroma and hues shifted toward UV wavelengths. Variation in feather color was best explained by sex, mass, parasite infection, and an interaction between area of oak scrub and group size. Juvenile feather color was also strongly associated with mother ID, but this was not the case for adults even though adult color was correlated with juvenile color. Adult chroma was reduced in birds treated with corticosterone, and hues were shifted toward longer wavelengths. No measure of adult color predicted acquisition of breeding space, but change in brightness and hue were significant predictors of acquisition for females, where breeders tended to experience greater reductions in brightness and shifts toward UV hues compared to helpers. This may be due to sex-specific differences in reproductive strategies in Florida Scrub-Jays. Females tend to disperse farther distances and breed earlier than males, potentially expending more energy foraying and searching for breeding space, which could reduce brightness if it is linked with condition. These results suggest that sexual selection is not a dominant factor in plumage ornamentation for Florida Scrub-Jays, and understanding the interaction between plumage color, personality, and reproductive fitness should be a priority for future research.

No abstract available. Copyright / Dissertation (MPhil)--University of Pretoria, 2011. / Taxation / unrestricted

No description available.

The k-nearest neighbour (k-NN) classifier is one of the oldest and most important supervised learning algorithms for classifying datasets. Traditionally the Euclidean norm is used as the distance for the k-NN classifier. In this thesis we investigate the use of alternative distances for the k-NN classifier. We start by introducing some background notions in statistical machine learning. We define the k-NN classifier and discuss Stone's theorem and the proof that k-NN is universally consistent on the normed space R^d. We then prove that k-NN is universally consistent if we take a sequence of random norms (that are independent of the sample and the query) from a family of norms that satisfies a particular boundedness condition. We extend this result by replacing norms with distances based on uniformly locally Lipschitz functions that satisfy certain conditions. We discuss the limitations of Stone's lemma and Stone's theorem, particularly with respect to quasinorms and adaptively choosing a distance for k-NN based on the labelled sample. We show the universal consistency of a two stage k-NN type classifier where we select the distance adaptively based on a split labelled sample and the query. We conclude by giving some examples of improvements of the accuracy of classifying various datasets using the above techniques.

A mobile robot needs to know its pose (position and orientation) in order to navigate and perform useful tasks. The problem of determining this pose with respect to a global or local frame is called localisation, and is a key component in providing autonomy to mobile robots. Thus, localisation answers the question Where am I? from the robot’s perspective. Localisation involving a single robot is a widely explored and documented problem in mobile robotics. The basic idea behind most documented localisation techniques involves the optimum combination of noisy and uncertain information that comes from various robot’s sensors. However, many complex robotic applications require multiple robots to work together and share information among themselves in order to successfully and efficiently accomplish certain tasks. This leads to research in collaborative localisation involving multiple robots. Several studies have shown that when multiple robots collaboratively localise themselves, the resulting accuracy in their estimated positions and orientations outperforms that of a single robot, especially in scenarios where robots do not have access to information about their surrounding environment. This thesis presents the main theme of most of the existing collaborative, multi-robot localisation solutions, and proposes an alternative or complementary solution to some of the existing challenges in multirobot localisation. Specifically, in this thesis, a heuristically tuned Extended Kalman Filter is proposed to localise a group of mobile robots. Simulations show that when certain conditions are met, the proposed tuning method significantly improves the accuracy and reliability of poses estimated by the Extended Kalman Filter. Real world experiments performed on custom-made robotic platforms validate the simulation results.

The convex hull of a sample is used to approximate the support of the underlying distribution. This approximation has many practical implications in real life. For example, approximating the boundary of a finite set is used by many authors in environmental studies and medical research. To approximate the functionals of convex hulls, asymptotic theory plays a crucial role. Unfortunately, the asymptotic results are mostly very complicated. To address this complication, we suggest a consistent bootstrapping scheme for certain cases. Our resampling technique is used for both semi-parametric and non-parametric cases. Let X1,X2,...,Xn be a sequence of i.i.d. random points uniformly distributed on an unknown convex set. Our bootstrapping scheme relies on resampling uniformly from the convex hull of X1,X2,...,Xn. In this thesis, we study the asymptotic consistency of certain functionals of convex hulls. In particular, we apply our bootstrapping technique to the Hausdorff distance between the actual convex set and its estimator. We also provide a conjecture for the application of our bootstrapping scheme to Gaussian polytopes. Moreover, some other relevant consistency results for the regular bootstrap are developed.

Mathematical conditions which showed where parsimony was not consistent for four taxa were first introduced by Felsenstein in 1978. This was subsequently labelled the "Felsenstein zone". Following Felsenstein's findings, 'frequentists' conjectured that for five taxa there would also be a region in parameter space where parsimony is not consistent. In response, 'cladists' claimed that parsimony was consistent in a different region of parameter space, which is called the "Farris zone". However, no analytical description of the region in which this consistency occurs has been made. Furthermore, no mathematical extensions of this Felsenstein theory to five taxa or more has been made. The same is true for the Farris zone. In this thesis, we give a complete account for the Felsenstein zone and Farris zone for four and five taxa and interpret these in terms of the shape of the phylogenetic tree.