Les réseaux véhiculaires sont des systèmes de communication basés sur un échange d'informations de véhicules à infrastructures fixes installées au bord des routes, on parle alors de mode V2I (Vehicle-to-Infrastructure), ou de véhicules à véhicules dit mode V2V (Vehicle-to-Vehicle) ou VANET (Vehicular Ad hoc Network). L'objectif est de fournir aux conducteurs et aux opérateurs de transport des informations sur le trafic routier permettant d'améliorer l'efficacité des systèmes de transport, la sécurité et le confort des usagers. Depuis leur apparition, les VANET ont connu un très grand essor, de nombreux standards, applications et mécanismes de routage ont été proposés pour répondre aux spécificités de cette nouvelle classe de réseaux. Les défis à relever pour leur conception découlent principalement de la forte mobilité des véhicules, de la diversité spatio-temporelle de la densité du trafic et de la propagation des ondes radio en environnement extérieur défavorable à l'établissement des communications sans fil. La difficulté, aussi bien économique que logistique, de la mise en œuvre réelle des réseaux véhiculaires fait de la simulation le moyen le plus largement utilisé pour la conception et l'évaluation des solutions proposées. Cependant la validité des résultats de simulation dépend fortement de la capacité des modèles utilisés à reproduire le plus fidèlement possible les situations réelles. Deux aspects sont essentiellement importants dans les VANET : la mobilité des véhicules et la propagation des ondes radio. Nous proposons dans cette thèse un nouveau modèle de mobilité et un nouveau modèle de propagation d’ondes radio pour réseaux de véhicules en environnement urbain et suburbain. Pour définir des schémas réalistes, ces deux modèles se basent sur des données statiques et dynamiques réelles sur les caractéristiques topographiques et socio-économiques de l'environnement. Ces données décrivent particulièrement la distribution spatio-temporelle des véhicules et les infrastructures présentes dans l'environnement. Trois cas d'études sont présentés dans la thèse pour la validation des modèles développés ; un environnement théorique, urbain ou suburbain, défini par l'utilisateur, notamment le cas Manhattan très utilisé, et deux environnements réels qui représentent des agglomérations de taille moyenne. Une autre contribution de cette thèse est l'étude de la connectivité radio et des performances des protocoles de routage dans les VANET. A partir de graphes dynamiques de connexions représentant la variation des liens radio entre véhicules en déplacement, nous avons analysé et déterminé les propriétés de la topologie des liaisons radio des réseaux véhiculaires. Pour étudier les protocoles de routage, nous avons utilisé le modèle de mobilité et le modèle de propagation radio que nous avons développés en association avec le simulateur de réseaux ns-2. Nous avons comparé les performances des protocoles de routage les plus répandus et déterminé les mécanismes de routage les plus adaptés aux réseaux véhiculaires. / Vehicular networks are communication systems based on information exchange either between vehicles and roadside fixed infrastructure, which is called V2I (Vehicle-to-Infrastructure) mode, or from vehicle to vehicle V2V (Vehicle-to-Vehicle) mode also known as VANET (Vehicular Ad hoc Network). The objective of these networks is to provide drivers and transport authorities the most timely information on road traffic in order to improve the efficiency of transportation systems, users safety and comfort.Since their appearance, the VANET have been greatly developed; many standards, applications and routing mechanisms have been proposed to address the specifics of this new class of networks. The challenges arise mainly from the high vehicles mobility, the spatiotemporal diversity of traffic density and, the radio waves propagation in external environment unfavorable to wireless communications establishment.The difficulty, both economic and logistical, of a real implementation of vehicular networks makes the simulation widely used to conceive and assess the proposed solutions. The validity of simulation results depends strongly on the ability of the models to reproduce as faithfully as possible the real situations. Two aspects are mainly important in the VANET: the simulation of vehicles mobility and radio wave propagation.We propose in this thesis a new mobility model and a new radio propagation model for vehicular networks in urban and suburban environment. To be realistic, these two models are based on real static and dynamic data describing the topographic and socioeconomic characteristics of the environment. These data concern particularly the spatiotemporal vehicles distribution and the description of the infrastructures present in the environment. Three case studies are presented in the thesis to validate the models, a theoretical user-defined urban or suburban environment (the Manhattan case very often used) and two real environments from mean size cities.Another contribution of this thesis is the study of radio connectivity and performance of routing protocols in the VANET. From dynamic graphs representing the variation of the radio links between vehicles in motion, we have analyzed and determined the topology properties of vehicular networks. To study routing protocols, we used the mobility model and the radio propagation model in association with the network simulator ns-2. We have compared the performance of the widespread routing protocols and determined the most adapted routing mechanisms to vehicular networks.

[Truncated abstract] The extent to which marine populations are connected by larval dispersal is crucial to their distribution, maintenance and diversity. Thus, for the effective conservation of threatened systems such as coral reefs, understanding patterns of connectivity is essential. However, the biophysical mechanisms that retain or disperse larvae within and among populations are poorly understood. Though the open ocean environment provides the opportunity for long-distance dispersal, if this potential is only rarely realised, recruits produced from afar are unlikely to contribute to the local-scale demography of populations over ecological time frames, but will limit broad-scale genetic diversification over evolutionary time. This thesis explores the extent of genetic and demographic connectivity of two species of reef-building corals over a range of spatial scales among the discontinuous reef systems of northwest Australia. ... Putative source and sink dynamics were not random, but were associated with levels of disturbance and recovery from a recent and catastrophic coral bleaching. When S. hystrix samples from another two offshore systems were included in the analysis, large differences among systems showed that gene flow over hundreds of kilometres is rare over microevolutionary time scales that account for connections over multiple generations. Levels of subdivision over the same spatial scales were markedly lower in the acroporid coral, Acropora tenuis, than in S. hystrix. These results are congruent with expectations based on reproductive mode; in contrast to S. hystrix, which releases brooded larvae that are competent to settle immediately, A. tenuis broadcasts its gametes, and after external fertilisation, the larvae need to develop for several days before they are competent to settle. Despite the differences in levels of broad-scale subdivision, in both species significant differentiation was detected between reefs within systems (>10 km), and between sites within some reefs (< 10 km). These results indicate not only that dispersal between reefs and even some reef patches is restricted, but also that hydrodynamics influence retention of brooded and spawned larvae in similar ways. Further analysis of A. tenuis populations from two coastal systems detected significant differences in genetic diversity among the four major systems of northwest Australia. Additionally, genetic divergence between the coastal and offshore zones was greater than expected by the geographic separation of systems, indicating that connectivity between these zones via transport of A. tenuis larvae on oceanic currents occurs rarely even over microevolutionary time scales. This study has two primary implications for conservation. First, since coastal and offshore reefs of northwest Australia appear to be discrete genetic entities, they have independent evolutionary potential to adapt to local conditions and environmental change. Second, systems, reefs and some reef patches of northwest Australia are demographically independent units. Therefore, designs of coral reserve networks should consider routine dispersal distances of kilometres to a few tens of kilometres.

<p>Increasingly wildlife managers and land managers are challenged to maintain the viability and connectivity among large mammal populations. Thus, it is important that effective highway crossings are identified and optimized with respect to construction cost, facilitation of ungulate movements, and ability to reduce wildlife-vehicle collisions. The use of exclusion fencing to reduce ungulate-vehicle collisions is commonly installed along Swedish highways. However, exclusion fences may pose a threat to the viability of wildlife populations because they serve as barriers to individual movements and may limit accessibility to resources. Various types of wildlife crossings intended to reduce road-kills and increase habitat connectivity across fenced highways have been constructed throughout the world. Previous studies have evaluated the importance of structure design and size for many ungulate species, but few studies involved moose (Alces alces) which is the target species for most large ungulate mitigations in Sweden. The results of the studies are intended to facilitate the development of wildlife crossings and conventional road passages to meet ungulate demands.</p><p>We monitored moose fitted with GPS radio collars and characterized their space and habitat use patterns in southwestern Sweden. Moose had seasonal differences in habitat selection within their home range, and generally preferred clear-cuts and early successional forest, boreal forest, and glades, but avoided agricultural areas and open water. We used infrared remote cameras, track surveys, and GPS telemetry to monitor the use of wildlife crossings and conventional road passages by moose and roe deer (Capreolus capreolus). The upgrading of a non-fenced road to a fenced highway with three wildlife crossings decreased the moose movements across the highway by 67-89 %. Overpass use by moose and roe deer declined with increased traffic volume on the highway and both species walked during periods of low highway traffic volumes and shifted to trotting as traffic intensity increased. Low rates of human disturbances and proximity to forest edges increased use of highway underpasses by roe deer. Moose used large underpasses to a higher degree than small.</p>

This thesis explores the relationship between crime, transportation network connectivity, and engineered facilities built to increase walkability through a case study of twenty-five sites within the city of Spokane, Washington. Using data spanning between 2008 and 2010, the author developed a GIS model representing conditions of the built environment, social and economic demographics, and crime rates within the randomly chosen sites. A partial F-test revealed little to no evidence that the transportation network connectivity affected crime rates within the study sites. Using permutation tests, the author found that specific pedestrian facility infrastructure correlated to higher crime rates as defined for the specific study sites. In all, this research provided patterns between crime rates and transportation networks which may be further explored in future research. / Graduation date: 2012

Increasingly wildlife managers and land managers are challenged to maintain the viability and connectivity among large mammal populations. Thus, it is important that effective highway crossings are identified and optimized with respect to construction cost, facilitation of ungulate movements, and ability to reduce wildlife-vehicle collisions. The use of exclusion fencing to reduce ungulate-vehicle collisions is commonly installed along Swedish highways. However, exclusion fences may pose a threat to the viability of wildlife populations because they serve as barriers to individual movements and may limit accessibility to resources. Various types of wildlife crossings intended to reduce road-kills and increase habitat connectivity across fenced highways have been constructed throughout the world. Previous studies have evaluated the importance of structure design and size for many ungulate species, but few studies involved moose (Alces alces) which is the target species for most large ungulate mitigations in Sweden. The results of the studies are intended to facilitate the development of wildlife crossings and conventional road passages to meet ungulate demands. We monitored moose fitted with GPS radio collars and characterized their space and habitat use patterns in southwestern Sweden. Moose had seasonal differences in habitat selection within their home range, and generally preferred clear-cuts and early successional forest, boreal forest, and glades, but avoided agricultural areas and open water. We used infrared remote cameras, track surveys, and GPS telemetry to monitor the use of wildlife crossings and conventional road passages by moose and roe deer (Capreolus capreolus). The upgrading of a non-fenced road to a fenced highway with three wildlife crossings decreased the moose movements across the highway by 67-89 %. Overpass use by moose and roe deer declined with increased traffic volume on the highway and both species walked during periods of low highway traffic volumes and shifted to trotting as traffic intensity increased. Low rates of human disturbances and proximity to forest edges increased use of highway underpasses by roe deer. Moose used large underpasses to a higher degree than small.

In this thesis, various control problems originating from the field of mobile robotics are considered. In particular, the thesis deals with problems that are related to the interaction and coordination of multiple mobile units. The scientific contributions are presented in five papers that together constitute the main part of the thesis. The papers are preceded by a longer introductory part, in which some important results from control theory, data processing and robotics are reviewed. In the first of the appended papers, two stabilizing tracking controls are proposed for a non-holonomic robot platform of unicycle type. Tolerance to errors and other properties of the controllers are discussed and a reactive obstacle avoidance control, that can easily be incorporated with the proposed tracking controls, is suggested. In Paper B, the results from Paper~A are extended to multi-agent systems. It is demonstrated how the tracking controls from Paper A can be used as building blocks when putting together formations of robots, in which each robot maintains a fixed position relative its neighbors during translation. In addition, switching between the different control functions is shown to be robust, implying that it is possible to change the shape of a formation on-line. In the first two papers, the tracking problem is facilitated by the assumption that the approximate velocity of the target/leader is known to the tracking robot. Paper C treats the the case where the target velocity is neither directly measurable with the available sensor setup, nor possible to obtain through communication with neighboring agents. Straight-forward computation of the target velocity from available sensor data unfortunately tend to enhance measurement errors and give unreliable estimates. To overcome the difficulties, an alternative approach to velocity estimation is proposed, motivated by the local observability of the given control system. Paper D deals with another problematic aspect of data acquisition. When using range sensors, one often obtains a mixed data set with measurements originating from many different sources. This problem would, for instance, be encountered by a robot moving in a formation, where it was surrounded by other agents. There exist established techniques for sorting mixed data sets off-line, but for time-depending systems where data need to be sorted on-line and only small time delays can be tolerated, established methods fail. The solution presented in the paper is a prediction-correction type algorithm, referred to as CCIA (Classification Correction and Identification algorithm). Finally, in Paper E, we consider the problem of maintaining connectivity in a multi-agent system. Often inter-agent communication abilities are associated with some proximity constraints, so when the robots move in relation to each other, communication links both break and form. In the paper we present a framework for analysis that makes it possible to compute a set of general constraints which, if satisfied, are sufficient to guarantee maintained communication for a given multi-agent system. Constraints are computed for two sorts of consensus-based systems and the results are verified in simulations. / QC 20100715

Functional Magnetic Resonance Imaging (fMRI) is one of the best techniques for neuroimaging and has revolutionized the way to understand the brain functions. It measures the changes in the blood oxygen level-dependent (BOLD) signal which is related to the neuronal activity. Complexity of the data, presence of different types of noises and the massive amount of data makes the fMRI data analysis a challenging one. It demands efficient signal processing and statistical analysis methods.  The inference of the analysis is used by the physicians, neurologists and researchers for better understanding of the brain functions.      The purpose of this study is to design a toolbox for fMRI data analysis. It includes methods to detect the brain activity maps, estimation of the hemodynamic response (HDR) and the connectivity of the brain structures. This toolbox provides methods for detection of activated brain regions measured with Bayesian estimator. Results are compared with the conventional methods such as t-test, ordinary least squares (OLS) and weighted least squares (WLS). Brain activation and HDR are estimated with linear adaptive model and nonlinear method based on radial basis function (RBF) neural network. Nonlinear autoregressive with exogenous inputs (NARX) neural network is developed to model the dynamics of the fMRI data.  This toolbox also provides methods to brain connectivity such as functional connectivity and effective connectivity.  These methods are examined on simulated and real fMRI datasets.

In neuroscience and biomedical engineering fields, one of the most important issues nowadays is finding a relationship between different brain regions when it is stimulated. Connectivity is an important research area in neuroscience which tries to determine the relationship between different brain region when the brain is stimulated externally or internally. Three main type of connectivity are discussed in this field: Anatomical, Functional and Effective connectivity. Importance of effective connectivity is its ability to detect brain disorders in early stages. Some brain disorders are Schizophrenia, MS and Major Depression disease. Comparing the effective connectivity between a healthy and unhealthy brain will help to diagnose brain disorder. In this master study, two methods named Dynamic Causal Modeling (DCM) and Psychophysiological Interaction (PPI) are used to compare effective connectivity and neuronal activity between different regions of brain when there are three different stimulations. Since the neural activity is latent in fMRI data, there is a need to a model which is able to transfer data from neuronal level to a visible data like Blood-Oxygen level dependent (BOLD) signal. DCM uses a haemodynamic balloon model (HD) to represent this data transfer. The hemodynamic model must be so that the parameters of neural and BOLD signal be the same. It should be noted that what is looked for is not the BOLD signal but the neuronal activity. In this study, as the first step, we did preprocessing of MR images and after ROI`s are created using the program MARSBAR. Ten ROIs, which are thought to have connections between them are selected by considering the stimulations used in the experiments in obtaining the data used in this thesis. The data used contains fMRI images of 11 healthy subjects. Stimulations of experiment are applied to images got from group analysis of 11 healthy subjects. These Stimulations are then used in preparing the design matrix and the parameters related to DCM. These parameters are the values related to connection matrices defining bilinear dynamic model on ROI. Bayesian method is used to select best model between all these models. Another method of PPI is also applied to analyze effective connectivity between 10 ROIs. This method considers two issues of physiological and psychological effects. Like DCM, the preprocessing steps and ROI selection is done for PPI and hemodynamic model is designed for this method. Neural and hemodynamic responses of ROIs are compared using this method.

This study sought to determine how five neighbourhood models (the Grid, Loop and Cul-de-Sac, Fused Grid, New Urbanist and Greenway) compare in terms of the characteristics known to affect active transportation rates, and which model is most likely to be able to facilitate active transportation as a result. In order to do so, model principles and design characteristics of case study neighbourhoods were described and used to create a range of design specifications for each model. These specifications were then used to develop a GIS-based representation of an example neighbourhood for each model, which included the transportation network, parcels of different land use types and densities, homes and destinations. GIS, statistical and graph-based techniques were then used to comprehensively assess and compare the models in terms of their potential to facilitate walking and biking through the built environment correlates identified in through a literature review. The models were ranked on each variable, and then an overall comparison was made on the basis diversity (land use mix), density and design - the three dimensions identified by Cervero and Kockelman (1997) as being the key ways through which the built environment can contribute to creating walkable (and potentially bikeable) neighbourhoods. Additional measures related to trip characteristics and issues of importance to developers (such as buildable area) were also included. The results illustrate how each model’s unique approach to facilitating walking and/or biking is reflected in the built environment characteristics assessed. While a model that was strong in one category was often weaker in another (a finding which echoes that of Filion and Hammond, 2003), the three alternative models (Fused Grid, New Urbanist and Greenway) consistently fared better than the more traditional Grid and Loop and Cul-de-Sac designs, with the New Urbanist scoring the highest on the overall evaluation of walkability and bikeability and the Greenway the best on network design for cyclists. In addition to these findings, the study also provided an opportunity to explore several challenges related to model assessment, such as issues arising from frame choice, off-set networks, and the use of roads as proxies for active transportation networks.

Angiography is an invasive procedure since contrast medium is injected into circulatory system of patients and the mostly preferred technique is X-ray angiography. For diagnosis, treatment planning, and risk assessment purposes, interventional radiologists utilize visual inspection to determine connectivity relations between vessels. This situation leads angiography to be more invasive, since it requires additional injection of contrast medium and X-ray dose. This thesis work presents a 3-D vascular connectivity tracking toolkit for automated extraction of vascular networks in 3-D medical images. The proposed method automatically extracts the vascular network connected to a user-defined point in a user-defined direction, and requires no further user interaction. The toolkit prevents additional injection of contrast agent and X-ray dose, saves time for the interventional radiologist. While the algorithm is applicable on all 3-D angiography images, performance of the method is observed on 3-D catheter angiography image of cerebrovascular structures. The algorithm iteratively tracks gravity centers of vascular branches in the user-defined direction, preserving connection to the user-defined point. Curvy branches are tracked even if they have discontinuous portions. Since this tracking method does not depend on lumen diameter and intensity differences, branches with stenoses and branches having large intensity difference can be successfully tracked. Skeletonization and junction detection methods are described, which are used to detect the sub branches, indirectly connected to the point. These methods are capable of handling bifurcations, trifurcations, and junctions having more branches. However, false junctions occurring due to superposition of vessels are not eliminated.