L’indexation par le contenu de lifelogs issus de capteurs portées a émergé comme un enjeu à forte valeur ajoutée permettant l’exploitation de ces nouveaux types de donnés. Rendu plus accessible par la récente disponibilité de dispositifs miniaturisés d’enregistrement, les besoins pour l’extraction automatique d’informations pertinents générées par autres applications, la localisation en environnement intérieur est un problème difficile à l’analyse de telles données.Beaucoup des solutions existantes pour la localisation fonctionnent insuffisamment bien ou nécessitent une intervention important à l’intérieur de bâtiment. Dans cette thèse, nous abordons le problème de la localisation topologique à partir de séquences vidéo issues d’une camera portée en utilisant une approche purement visuelle. Ce travail complète d’extraction des descripteurs visuels de bas niveaux jusqu’à l’estimation finale de la localisation à l’aide d’algorithmes automatiques.Dans ce cadre, les contributions principales de ce travail ont été faites pour l’exploitation efficace des informations apportées par descripteurs visuels multiples, par les images non étiquetées et par la continuité temporelle de la vidéo. Ainsi, la fusion précoce et la fusion tardive des données visuelles ont été examinées et l’avantage apporté par la complémentarité des descripteurs visuels a été mis en évidence sur le problème de la localisation. En raison de difficulté à obtenir des données étiquetées en quantités suffisantes, l’ensemble des données a été exploité ; d’une part les approches de réduction de dimensionnalité non-linéaire ont été appliquées, afin d’améliorer la taille des données à traiter et la complexité associée ; d’autre part des approches semi-supervisés ont été étudiées pour utiliser l’information supplémentaire apportée par les images non étiquetées lors de la classification. Ces éléments ont été analysé séparément et on été mis en œuvre ensemble sous la forme d’une nouvelle méthode par co-apprentissage temporelle. Finalement nous avons également exploré la question de l’invariance des descripteurs, en proposant l’utilisation d’un apprentissage invariant à la transformation spatiale, comme un autre réponse possible un manque de données annotées et à la variabilité visuelle.Ces méthodes ont été évaluées sur des séquences vidéo en environnement contrôlé accessibles publiquement pour évaluer le gain spécifique de chaque contribution. Ce travail a également été appliqué dans le cadre du projet IMMED, qui concerne l’observation et l’indexation d’activités de la vie quotidienne dans un objectif d’aide au diagnostic médical, à l’aide d’une caméra vidéo portée. Nous avons ainsi pu mettre en œuvre le dispositif d’acquisition vidéo portée, et montrer le potentiel de notre approche pour l’estimation de la localisation topologique sur un corpus présentant des conditions difficiles représentatives des données réelles. / Visual lifelog indexing by content has emerged as a high reward application. Enabled by the recent availability of miniaturized recording devices, the demand for automatic extraction of relevant information from wearable sensors generated content has grown. Among many other applications, indoor localization is one challenging problem to be addressed.Many standard solutions perform unreliably in indoors conditions or require significant intervention. In this thesis we address from the perspective of wearable video camera sensors using an image-based approach. The key contribution of this work is the development and the study of a location estimation system composed of diverse modules, which perform tasks ranging from low-level visual information extraction to final topological location estimation with the aid of automatic indexing algorithms. Within this framework, important contributions have been made by efficiently leveraging information brought by multiple visual features, unlabeled image data and the temporal continuity of the video.Early and late data fusion were considered, and shown to take advantage of the complementarities of multiple visual features describing the images. Due to the difficulty in obtaining annotated data in our context, semi-supervised approaches were investigated, to use unlabeled data as additional source of information, both for non-linear data-adaptive dimensionality reduction, and for improving classification. Herein we have developed a time-aware co-training approach that combines late data-fusion with the semi-supervised exploitation of both unlabeled data and time information. Finally, we have proposed to apply transformation invariant learning to adapt non-invariant descriptors to our localization framework.The methods have been tested on controlled publically available datasets to evaluate the gain of each contribution. This work has also been applied to the IMMED project, dealing with activity recognition and monitoring of the daily living using a wearable camera. In this context, the developed framework has been used to estimate localization on the real world IMMED project video corpus, which showed the potential of the approaches in such challenging conditions.

Many wireless sensor network (WSN) applications depend on knowing the position of nodes within the network if they are to function efficiently. Location information is used, for example, in item tracking, routing protocols and controlling node density. Configuring each node with its position manually is cumbersome, and not feasible in networks with mobile nodes or dynamic topologies. WSNs, therefore, rely on localisation algorithms for the sensor nodes to determine their own physical location. The basis of several localisation algorithms is the theory that the higher the number of reference nodes (called “references”) used, the greater the accuracy of the estimated position. However, this approach makes computation more complex and increases the likelihood that the location estimation may be inaccurate. Such inaccuracy in estimation could be due to including data from nodes with a large measurement error, or from nodes that intentionally aim to undermine the localisation process. This approach also has limited success in networks with sparse references, or where data cannot always be collected from many references (due for example to communication obstructions or bandwidth limitations). These situations require a method for achieving reliable and accurate localisation using a limited number of references. Designing a localisation algorithm that could estimate node position with high accuracy using a low number of references is not a trivial problem. As the number of references decreases, more statistical weight is attached to each reference’s location estimate. The overall localisation accuracy therefore greatly depends on the robustness of the selection method that is used to eliminate inaccurate references. Various localisation algorithms and their performance in WSNs were studied. Information-fusion theory was also investigated and a new technique, rooted in information-fusion theory, was proposed for defining the best criteria for the selection of references. The researcher chose selection criteria to identify only those references that would increase the overall localisation accuracy. Using these criteria also minimises the number of iterations needed to refine the accuracy of the estimated position. This reduces bandwidth requirements and the time required for a position estimation after any topology change (or even after initial network deployment). The resultant algorithm achieved two main goals simultaneously: accurate location discovery and information fusion. Moreover, the algorithm fulfils several secondary design objectives: self-organising nature, simplicity, robustness, localised processing and security. The proposed method was implemented and evaluated using a commercial network simulator. This evaluation of the proposed algorithm’s performance demonstrated that it is superior to other localisation algorithms evaluated; using fewer references, the algorithm performed better in terms of accuracy, robustness, security and energy efficiency. These results confirm that the proposed selection method and associated localisation algorithm allow for reliable and accurate location information to be gathered using a minimum number of references. This decreases the computational burden of gathering and analysing location data from the high number of references previously believed to be necessary. / Thesis (PhD(Eng))--University of Pretoria, 2011. / Electrical, Electronic and Computer Engineering / unrestricted

To solve the spectrum scarcity problem caused by the high number of wireless applications and users, the concept of cognitive radio (CR) was proposed in the past few years. Cognitive radio networks (CRNs) provide dynamic spectrum access (DSA), where the unlicensed users can access the spectrum without causing unacceptable level of interference to the primary user (PU). DSA was based on conventional spectrum sensing information or geolocation databases. Later, radio environment map (REM) as an improved geolocation database was introduced to enhance the DSA process. It is a comprehensive map consists of different integrated databases, and the interference field information is one of its databases. In this thesis, a description of the REM concept and its construction methods will be presented. The focus will be for the indirect methods for constructing interference map, which represents a layer of the REM. Indirect method refers to the methods that utilize known model information, to first estimate the primary transmitter parameters and then generate REM. Two indirect methods under lognormal shadowing were presented and compared. The better of these two methods is further investigated in different scenarios. These scenarios include different number of sensors, varied size of measurements, several shadowing spread values, different percentages of error in path-loss exponent, and the effect of the number of moving sensors and their speeds to the REM quality. The performance is evaluated using these metrics: “localization error, signal power error and correct detection zone ratio (CDZR). The results show that performance is enhanced as the number of sensors and the size of measurements increase, whereas clear degradation in REM quality is shown when shadowing spread increases or the model parameters are not well calibrated. Also, as the number of moving sensors or their speeds increase, the REM performance becomes less effective

To θέμα του εντοπισμού των κινητών τηλεφώνων έχει τραβήξει την προσοχή τα τελευταία χρόνια εξαιτίας των απαιτήσεων της Ομοσπονδιακής Επιτροπής Επικοινωνιών για το Enhanced 911 η οποία είναι μια υπηρεσία συναγερμού. Τα ασύρματα συστήματα επικοινωνίας 3ης γενιάς ηταν τα πρώτα που υιοθέτησαν στρατηγικές εύρεσης θέσης στα στάνταρντ τους. Στην διαδικασία της εύρεσης της θέσης υπάρχουν 3 βασικές κατηγορίες μετρήσεων που μπορούμε να χρησιμοποιήσουμε. Η πρώτη εκτιμά το κινητό βασίζοντας τις μετρήσεις στην λαμβανόμενη ισχύ σήματος. Η δεύτερη κάνει χρήση των χρόνων άφιξης ή της διαφοράς των χρόνων άφιξης στους σταθμούς βάσης. Η τρίτη κατηγορία έχει να κάνει με τις γωνίες άφιξης στους σταθμούς βάσης. Όλες οι παραπάνω κατηγορίες μετρήσεων υποβαθμίζονται έντονα από την NLOS διάδοση. Η απουσία ενός LOS μονοπατιού μπορεί να βλάψει σημαντικά την εκτίμηση της πραγματικής θέσης του κινητού. Αντικείμενο της παρούσας διπλωματικής είναι η συγκριτική αξιολόγηση και μελέτη κάποιων δημοφιλών τεχνικών εντοπισμού θέσης απλών και υβριδικών κάτω από διαφορετικά ΝLOS περιβάλλοντα και σε συνδυασμό με άλλες εξίσου σημαντικές παραμετρους όπως ειναι το διαθέσιμο πλήθος σταθμών βάσης σε μια περιοχή, η γεωμετρία ή με άλλα λόγια η θέση του κινητού σε σχέση με τους σταθμους βάσης. Προτείνεται επίσης και μια υβριδική τεχνική για την αντιμετώπιση των παραπάνω καταστρεπτικών επιπτώσεων του NLOS φαινομένου. / The problem of mobile location estimation has recently drawn attention due to Federal Communications Commission (FCC) demands of Enhanced-911 (E911) emergency service. Third Generation (3G) wireless systems were the first to adopt location estimation techniques into their standards. There are three basic types of measurements that can be used for location estimation. The first type includes Received Signal Strength measurements. The second type uses Time of Arrival or Time Difference of Arrival measurements of the signal to the base stations. The third type deals with Angle of Arrival measurements of the received signal. The subject of this work is the comparative evaluation and study of certain popular, simple and hybrid location estimation techniques, under different NLOS environments and in conjunction with other equally important parameters such as the number of available base stations, the geometry of the problem and the position of the mobile relative to the base stations. A hybrid method is also suggested for mitigating the destructive consequences of the NLOS effect.

An iterative travel time forecasting scheme, named the Advanced Multilane Prediction based Real-time Fastest Path (AMPRFP) algorithm, is presented in this dissertation. This scheme is derived from the conventional kernel estimator based prediction model by the association of real-time nonlinear impacts that caused by neighboring arcs’ traffic patterns with the historical traffic behaviors. The AMPRFP algorithm is evaluated by prediction of the travel time of congested arcs in the urban area of Jacksonville City. Experiment results illustrate that the proposed scheme is able to significantly reduce both the relative mean error (RME) and the root-mean-squared error (RMSE) of the predicted travel time. To obtain high quality real-time traffic information, which is essential to the performance of the AMPRFP algorithm, a data clean scheme enhanced empirical learning (DCSEEL) algorithm is also introduced. This novel method investigates the correlation between distance and direction in the geometrical map, which is not considered in existing fingerprint localization methods. Specifically, empirical learning methods are applied to minimize the error that exists in the estimated distance. A direction filter is developed to clean joints that have negative influence to the localization accuracy. Synthetic experiments in urban, suburban and rural environments are designed to evaluate the performance of DCSEEL algorithm in determining the cellular probe’s position. The results show that the cellular probe’s localization accuracy can be notably improved by the DCSEEL algorithm. Additionally, a new fast correlation technique for overcoming the time efficiency problem of the existing correlation algorithm based floating car data (FCD) technique is developed. The matching process is transformed into a 1-dimensional (1-D) curve matching problem and the Fast Normalized Cross-Correlation (FNCC) algorithm is introduced to supersede the Pearson product Moment Correlation Co-efficient (PMCC) algorithm in order to achieve the real-time requirement of the FCD method. The fast correlation technique shows a significant improvement in reducing the computational cost without affecting the accuracy of the matching process.

This study selects several new advanced topics in power systems, and verifies their usefulness using the simulation. In the study on ratio of the equivalent reactance and resistance of the bulk power systems, the simulation results give us the more correct value of X/R of the bulk power system, which can explain why the active power compensation is also important in voltage flicker mitigation. In the application study of the Energy Storage System(ESS) to the wind power, the new model implementation of the ESS connected to the wind power is proposed, and the control effect of ESS to the intermittency of the wind power is verified. Also this study conducts the intensive simulations for clarifying the behavior of the wide-area power system frequency as well as the possibility of the on-line instability detection. In our POWER IT Laboratory, since 2003, the U.S. national frequency monitoring network (FNET) has been being continuously operated to monitor the wide-area power system frequency in the U.S. Using the measured frequency data, the event of the power system is triggered, and its location and scale are estimated. This study also looks for the possibility of using the simulation technologies to contribute the applications of FNET, finds similarity of the event detection orders between the frequency measurements and the simulations in the U.S. Eastern power grid, and develops the new methodology for estimating the event location based on the simulated N-1 contingencies using the frequency measurement. It has been pointed out that the simulation results can not represent the actual response of the power systems due to the inevitable limit of modeling power systems and different operating conditions of the systems at every second. However, in the circumstances that we need to test such an important infrastructure supplying the electric energy without taking any risk of it, the software based simulation will be the best solution to verify the new technologies in power system engineering and, for doing this, new models and better application of the simulation should be proposed. Conducting extensive simulation studies, this dissertation verified that the actual X/R ratio of the bulk power systems is much lower than what has been known as its typical value, showed the effectiveness of the ESS control to mitigate the intermittence of the wind power from the perspective of the power grid using the newly proposed simulation model of ESS connected to the wind power, and found many characteristics of the wide-area frequency wave propagation. Also the possibility of using the simulated responses of the power system for replacing the measured data could be confirmed and this is very promising to the future application of the simulation to the on-line analysis of the power systems based on the FNET measurements. / Ph. D.

Wide Area Measurements (WAMs) have been widely used in the energy management system (EMS) of power system for monitoring, operation and control. In recent years, the advent of synchronized Phasor Measurements Unit (PMU) has added another dimension to the field of wide-area measurement. However, the high cost of the PMU, which includes the manufacture and deployment fee, is a hurdle to the wide use of the PMU in power systems. Unlike traditional PMUs, the frequency monitoring network (FNET) developed by the Virginia Tech Power IT lab is an Internet—based, GPS—synchronized, wide-area frequency monitoring network deployed at the distribution level, providing a low-cost and easily deployable WAMs solution. In this dissertation, the research work can be categorized into two parts: FNET Design and Situation Awareness Algorithm Development. / Ph. D.

The rapid growth of the electric power system has in recent decades resulted in an increase of the number of transmission lines and total power outage in Norway. The challenge of a fast growing electrical grid has also resulted in huge increases of overhead lines and their total length. These lines are experiencing faults due to various reasons that cause major disruptions and operating costs of the transmission system operator (TSO). Thus, it’s important that the location of faults is either known or can be estimated with reasonably high accuracy. This allows the grid owner to save money and time for inspection and repair, as well as to provide a better service due to the possibility of faster restoration of power supply and avoiding blackouts.  Fault detection and classification on transmission lines are important tasks in order to protect the electrical power system. In recent years, the power system has become more complicated under competitive and deregulated environments and a fast fault location technique is needed to maintain security and supply in the grid. This thesis compares and evaluates different methods for classification of fault type and calculation of conventional one-side and two-side based fault location algorithms for distance to fault estimation.  Different algorithm has been implemented, tested and verified to create a greater understanding of determinants facts that affect distance to faults algorithm’s accuracy.  Implemented algorithm has been tested on the data generated from a number of simulations in Simulink for a verification process in implemented algorithms accuracy. Two types of fault cases have also been simulated and compared for known distance to fault estimation.