L’objectif de cette thèse est la classification automatique des cibles (ATC) en utilisant les signaux rétrodiffusés par un radar ultra large bande (UWB). La classification des cibles est réalisée en comparant les signatures des cibles et les signatures stockées dans une base de données. Premièrement, une étude sur la théorie de diffusion nous a permis de comprendre le sens physique des paramètres extraits et de les exprimer mathématiquement. Deuxièmement, des méthodes d’extraction de paramètres sont appliquées afin de déterminer les signatures des cibles. Un bon choix des paramètres est important afin de distinguer les différentes cibles. Différentes méthodes d’extraction de paramètres sont comparées notamment : méthode de Prony, Racine-classification des signaux multiples (Root-MUSIC), l’estimation des paramètres des signaux par des techniques d’invariances rotationnels (ESPRIT), et la méthode Matrix Pencil (MPM). Troisièmement, une méthode efficace de classification supervisée est nécessaire afin de classer les cibles inconnues par l’utilisation de leurs signatures extraites. Différentes méthodes de classification sont comparées notamment : Classification par la distance de Mahalanobis (MDC), Naïve Bayes (NB), k-plus proches voisins (k-NN), Machines à Vecteurs de Support (SVM). Une bonne technique de classification doit avoir une bonne précision en présence de signaux bruités et quelques soit l’angle d’émission. Les différents algorithmes ont été validés en utilisant les simulations des données rétrodiffusées par des objets canoniques et des cibles de géométries complexes modélisées par des fils minces et parfaitement conducteurs. Une méthode de classification automatique de cibles basée sur l’utilisation de la méthode Matrix Pencil dans le domaine fréquentiel (MPMFD) pour l’extraction des paramètres et la classification par la distance de Mahalanobis est proposée. Les résultats de simulation montrent que les paramètres extraits par MPMFD présentent une solution plausible pour la classification automatique des cibles. En outre, nous avons prouvé que la méthode proposée a une bonne tolérance aux bruits lors de la classification des cibles. Enfin, les différents algorithmes sont validés sur des données expérimentales et cibles réelles. / The objective of this thesis is the Automatic Target Classification (ATC) based on radar backscattered Ultra WideBand (UWB) signals. The classification of the targets is realized by making comparison between the deduced target properties and the different target features which are already recorded in a database. First, the study of scattering theory allows us to understand the physical meaning of the extracted features and describe them mathematically. Second, feature extraction methods are applied in order to extract signatures of the targets. A good choice of features is important to distinguish different targets. Different methods of feature extraction are compared including wavelet transform and high resolution techniques such as: Prony’s method, Root-Multiple SIgnal Classification (Root-MUSIC), Estimation of Signal Parameters via Rotational Invariance Techniques (ESPRIT) and Matrix Pencil Method (MPM). Third, an efficient method of supervised classification is necessary to classify unknown targets by using the extracted features. Different methods of classification are compared: Mahalanobis Distance Classifier (MDC), Naïve Bayes (NB), k-Nearest Neighbors (k-NN) and Support Vector Machine (SVM). A useful classifier design technique should have a high rate of accuracy in the presence of noisy data coming from different aspect angles. The different algorithms are demonstrated using simulated backscattered data from canonical objects and complex target geometries modeled by perfectly conducting thin wires. A method of ATC based on the use of Matrix Pencil Method in Frequency Domain (MPMFD) for feature extraction and MDC for classification is proposed. Simulation results illustrate that features extracted with MPMFD present a plausible solution to automatic target classification. In addition, we prove that the proposed method has better ability to tolerate noise effects in radar target classification. Finally, the different algorithms are validated on experimental data and real targets.

We study canonical forms for Hamiltonian and symplectic matrices or pencils under equivalence transformations which keep the class invariant. In contrast to other canonical forms our forms are as close as possible to a triangular structure in the same class. We give necessary and sufficient conditions for the existence of Hamiltonian and symplectic triangular Jordan, Kronecker and Schur forms. The presented results generalize results of Lin and Ho [17] and simplify the proofs presented there.

No description available.

Le travail de recherche proposé est relatif au suivi de la consommation par usage afin que les clients puissent mieux maîtriser leurs consommations électriques. Ce projet de recherche consiste à identifier les charges électriques en milieu domestique à partir des mesures non intrusives faites au niveau du compteur électrique. Les informations relevées et traitées sont les courants et les tensions. Nous avons proposé une méthode innovante et performante d'identification, c'est la méthode de Matrix Pencil. Elle s'adapte à la fois aux parties transitoires que permanentes des signatures traitées. Nous proposons deux voies d'investigations complémentaires en basses fréquences et en hautes fréquences auxquelles on appliquera la méthode de Matrix Pencil pour caractériser chaque charge par un ensemble de pôles et de résidus tant en régime transitoire qu'en régime permanent. Cette technique d'identification a été implantée dans le prototype de compteur électrique réalisant les fonctions de reconnaissance et d'identification des signatures. La conception et la réalisation du prototype ont été suivies par la validation de la fonction d'identification et de reconnaissance des usages dans une maison témoin. La capacité d'identification et le niveau du taux d'erreur sont très satisfaisants. Une éventuelle amélioration technologique permettra dans le futur de réaliser une meilleure identification.

Are computer-administered versions of a multiple choice paper-pencil test equivalent? This study determined whether there were any significant differences between taking a traditional pencil-paper test and taking the same test using a computer. The literature has shown that there are intervening variables that have caused differences when not controlled. To prove equivalency between test modes, scores have to have similar means, dispersions, and shapes; the ranked-order of the scores must also be similar. Four tests were given over the course of a 16-week semester. The sample was divided, half taking paper-pencil tests and half taking the same test administered by a computer. The mode of administration was switched with each test administration. The analysis showed that, when the intervening variables were controlled, the two modes of administration were equivalent. The analysis used a 2x4 ANOVA, which showed no difference between test modes, but showed that each test administration was significantly different. The Levene statistic was used to test whether dispersions were equivalent and confidence intervals were established to test the kurtosis and skewness statistics. Finally, each of the test scores were transformed into their Normal Curve Equivalents so that Pearson's coefficient could be used to determine the equivalency of the ranked-orders.

The problem of this study was to investigate the strategies used by high school students of limited estimation ability for the estimation of the answers to computational problems. The Assessing Computational Estimation Test was administered to 460 students, and 40 of them were selected for interviews. Each student interviewed was asked to estimate the answers to fourteen computation and application problems.

In this dissertation, a method for producing multinets from a net in P^3 is presented. Multinets play an important role in the study of resonance varieties of the complement of a complex hyperplane arrangement and very few examples are known. Implementing this method, numerous new and interesting examples of multinets are identified. These examples provide additional evidence supporting the conjecture of Pereira and Yuzvinsky that all multinets are degenerations of nets. Also, a complete description is given of proper weak multinets, a generalization of multinets.

The purpose of this study was to test whether the computer-administered College Level Examination Program (CLEP) General Examination in English Composition produced scores equivalent to those obtained from the traditional paper-and-pencil version. The CLEP examination and its adaptation for computer administration and the results of a pilot study are presented. The subjects in this study were volunteers who took the CLEP English Composition Examination in order to earn college credit and were randomly assigned to either the computer-first or paper-and-pencil-first groups. Each subject took both forms of the examination with approximately one half of the subjects taking each version first. Analysis of variance failed to detect a significant effect for test form or an interaction of test form and order of administration. Equivalence reliability coefficients and internal consistency coefficients also indicated that the computer administration did not significantly alter the results of the examination. In general, the results of this study support the hypothesis that the computer-administered version of the CLEP General Examination in English Composition produces results equivalent to those obtained from the traditional paper-and-pencil version.

Tropical African forests are thought to be declining due to human activity, resulting in serious threat to local communities and wildlife. The threatened status of populations of Juniperus procera and Olea africana was studied in Loliondo, Tanzania. Three hundred persons were interviewed in ten Maasai communities concerning tree utilization for fence and house construction, firewood, traditional medicine and gourd sterilization. By comparing the indices for 169 species, J. procera and O. africana usage was significantly higher than all other species (Kruskal-Wallis, chi2 = 15.3, DF = 2, p = 0.0005). Ninety-six percent of interviewees have witnessed local decline in J. procera and O. africana. In conjunction with interviews, ecological sampling was carried out in community forests to determine species status. The correlations of size class distributions are not typical of "healthy" plant populations. If current trends continue local extinction of J. procera and O. africana may occur.

国立情報学研究所で電子化したコンテンツを使用している。