In today’s technically advanced society the collection and study of digital images has become an important aspect of various off-line applications that range from medical diagnosis to exploring the Martian surface for traces of water. Various industries have recently started moving towards vision based systems to monitor several of their manufacturing processes. Except for some simple on-line applications, these systems are primarily used to analyze the digital images off-line. This thesis is concerned with developing a more powerful on-line digital image analysis technique which links the fields of traditional digital image processing with a recently devised statistically based image analysis method called multivariate image analysis (MIA). The first part of the thesis introduces the area of traditional digital image processing techniques through a brief literature review of three of its five main classes (image enhancement, restoration, analysis, compression, & synthesis) which contain most of the commonly used operations in this area. This introduction is intended as a starting point for readers who have little background in this field, and as a means of providing sufficient details on these techniques so that they can be used in conjunction with other advanced MIA on-line monitoring operations. MIA of multispectral digital images using latent variable statistical methods (Multi-Way PCA / PLS) is the main topic covered by the second part of this thesis. After reviewing the basic theory of feature extraction using MIA for off-line analyses, a new technique is introduced that extends these ideas for image analyses in on-line applications. Instead of directly using the updated images themselves to monitor a time- varying process, this new technique uses the latent variable space of the image to monitor the increase or decline in the number of pixels belonging to various features of interest. The ability to switch between the images and their latent variable space then allows the user to determine the exact spatial locations of any features of interest. This new method is shown to be ideal for monitoring interesting features from time-varying processes equipped with multispectral sensors. It forms a basis for future on-line industrial process monitoring schemes in those industries that are moving towards automatic vision systems using multispectral digital imagery. / Thesis / Master of Engineering (ME)

ITC/USA 2015 Conference Proceedings / The Fifty-First Annual International Telemetering Conference and Technical Exhibition / October 26-29, 2015 / Bally's Hotel & Convention Center, Las Vegas, NV / Power supply is the most critical unit in an airplane. A defect in a power supply might lead to a crash. Therefore, it is very useful and important to monitor the status of the airplane's power supply during flight. This paper presents a real-time evaluation system constructed by a central process unit for processing and detecting defects and several remote units for acquiring parameters. Because of the huge amount of data acquired, a combination of Ethernet and time division multiplexing technique has been applied for constructing a 2.5Gbps synchronous data transportation network for the system.

Most control systems now are assumed to be unchangeable, but this is an ideal situation. In real applications, they are often accompanied with many changes. Some of changes are from environment changes, and some are system requirements. So, the goal of thesis is to model a dynamic adaptive real-time control system process with big data stream. In this way, control system model can adjust itself using example measurements acquired during the operation and give suggestion to next arrival input, which also indicates the accuracy of states under control highly depends on quality of the process model.   In this thesis, we choose recurrent neural network to model process because it is a kind of cheap and fast artificial intelligence. In most of existent artificial intelligence, a database is necessity and the bigger the database is, the more accurate result can be. For example, in case-based reasoning, testcase should be compared with all of cases in database, then take the closer one’s result as reference. However, in neural network, it does not need any big database to support and search, and only needs simple calculation instead, because information is all stored in each connection. All small units called neuron are linear combination, but a neural network made up of neurons can perform some complex and non-linear functionalities. For training part, Backpropagation and Kalman filter are used together. Backpropagation is a widely-used and stable optimization algorithm. Kalman filter is new to gradient-based optimization, but it has been proved to converge faster than other traditional first-order-gradient-based algorithms.   Several experiments were prepared to compare new and existent algorithms under various circumstances. The first set of experiments are static systems and they are only used to investigate convergence rate and accuracy of different algorithms. The second set of experiments are time-varying systems and the purpose is to take one more attribute, adaptivity, into consideration.

Les dernières décennies ont connu l’apogée des transmissions hertziennes, et celles à venir connaîtront incontestablement le développement et le déploiement intense de systèmes de communications sans-fil. Dès à présent, il est possible de communiquer par onde sur petite et très petite distance (LAN et PAN). Les populations se sont familiariséesavec les interfaces bluetooth (IEEE802.15) présentes dans la majorité des objetscommuniquant (ordinateur portable, téléphone, PDA, etc...). Les foyers s’équipentmaintenant facilement et à bas prix d’interfaceWi-Fi (IEEE802.11), afin de profiter d’uneutilisation nomade de leur accès internet. Ainsi, la forte croissance dumarché des offresinternet combinée avec celle du marché des téléphones mobiles ont habitués un large spectre d’utilisateurs à communiquer sans fil. Ce contexte sociologique et financier encourage donc l’arrivée de solutions nouvelles répondant à des besoins latents. Parmi ceux-là, le marché met en évidence le manque de système de communication sur moyenne distance (MAN). Les réseaux ad-hoc peuvent répondre à ce genre de besoin. Mais àce jour, les performances sont trop faibles pour les besoins des utilisateurs et elles dépendenttrop fortement de la densité desmachines nomades. Aussi, le consortiumIEEEcherche au travers de sa norme IEEE802.16 à fournir un système complet de communicationsans-fil sur moyenne distance (MAN). Appelé aussiWiMAX, ce système se basesur une architecture composée d’une station de base (BS) et de nombreux mobiles utilisateurs(SS). Le standard IEEE802.16 définit les caractéristiques de la couche physiqueet de la couche MAC. Il décrit l’ensemble des interactions et événements pouvant avoirlieu entre la station de base et les stations mobiles. Enfin, le standard fournit différents paramètres et variables servant aux mécanismes de communication. Comme tout nouveau standard émergeant, la norme IEEE802.16 ne profite pas d’un état de l’art aussi développé que celui du IEEE802.11 par exemple. Aussi, de nombreuses études et idées sont à développer.En premier lieu, nous effectuons un large rappel de la norme WiMAX et en particulier le IEEE802.16e. Associé à cela, nous dressons un état de l’art des travaux traitant des aspects et perspectives liés au sujet de notre étude.Par la suite, nous proposons un modèle novateur de performance des communicationsIEEE802.16e. Au travers de ce modèle, nous développons une étude générale et exhaustive des principaux paramètres de communication. L’étude explicite l’impact deces paramètres ainsi que l’influence de leur évolutions possibles. De cela, nous critiquonsla pertinence de chacun d’eux en proposant des alternatives de configurations.5En sus, nous proposons un mécanisme novateur favorisant le respect de qualité de service(QoS) sur couche AC.Nous développons un principe original d’établissement de connexion favorisant l’accès aux communications sensibles aux délais de transmission.Dans une dernière partie, nous déterminons la capacité d’un système IEEE802.16 à gérer les arrivées et départs des utilisateurs. Tout en y associant une étude de performance d’un nouvel algorithme de contrôle d’admission. Cet algorithme d’admission vise à remplir des objectifs multiples : empêcher les famines de ressources sur les trafics les moins prioritaires, favoriser l’admission des utilisateurs en maintenant une gestion optimale de la ressource radio. Notre étude aboutit à une modélisation et une critique des variations de paramètre associés à ce nouvel algorithme. Nous y intégrons par la suite le principe de mobilité où les utilisateurs ont la capacité de se mouvoir au sein d’une cellule. Cette intégration se fait en y associant des mécanismes originaux afin d’assurer la pérennité du service aux utilisateurs mobiles. / The last decade witnessed the peak of the hertzian communications. The following ones will undoubtedly testify the intensive deployment and the development of all wireless ways of transmission. Due to cheaper equipments, the people are now used with all sorts of connected objects : laptop, smartphone, pad, and more recently, Connected Video display and audio diffusers. All these stuffs allow to keep an access to internet, even in a nomad use. This economical and sociological context promotes the emerging of new solutions metting latent needs by offering better performances. Consumer studies highlight particularly the lack of transmissions solution for Metropolitan Area Networks (MAN). Ad-hoc wireless solutions lead to satisfy the MAN needs, but the throughput is too importantly related to the ad-hoc customer capacity and density over the MAN coverage.The IEEE consortium seeks, through its IEEE802.16e standart, to provide a wireless transmission technology specifically design for the middle range network. Knowed as WiMAX, this system are based on a point to multipoint architecture. WiMAX standart gathers Base Station (BS) and Subscriber Stations (SS), and defines for both the Physical and MAC layer in the OSI Model. In addition, the standart proposes a set of default parameters for the two first OSI Layers.As any emerging standart, the IEEE802.16e suffers form a lack of litterature (works, studies and enhancement proposals). More studies are explicitly needed to craft and tune the IEEE802.16e standart in order to better answer to the specific issues met in the actual context of transmission.In a first step, we present in a large scale the IEEE802.16e standart specifications. In addition we highlight the main state of art linked to this subject.Second, we propose an original performance model, the first one that takes in account all the MAC layer parameters of the standart. Based on this model, we lead a general and exhaustive performance study of each communication parameters. This study highlights the importance of each parameters and propose some enhancements in fonction of the type of Quality of Service (QoS). In addition, we introduce an call engaging mecanism which respects the QoS on the MAC layer.In a last part, we compute the IEEE802.16e capacity to manage the incoming and leaving calls. we perform this study by introducing a new Connection Admission Control (CAC). The CAC algorithm achieves sevral objectives : prevent from the lack of ressource for the lowest priority flows as well as optimize the radio resource consumption to facilitate the access for the users. Our study is concluded by proposing an new capacity model and algorithm for the CAC. Moreover, this last proposal prevents the call drop due to user mobility.