<p> This thesis presents a novel hybrid methodology using Recurrent Neural Network and Dynamic Time Warping to solve the mode estimation problem of a radar warning receiver (RWR). The RWR is an electronic support (ES) system with the primary objective to estimate the threat posed by an unfriendly (hostile) radar in an electronic warfare (EW) environment. One such radar is the multi-function radar (MFR), which employs complex signal architecture to perform multiple tasks. As the threat posed by the radar directly depends on its current mode of operation, it is vital to estimate and track the mode of the radar. The proposed method uses a recurrent neural network (echo state network and recurrent multi-layer perceptron) trained in a supervised manner, with the dynamic time warping algorithm as the post processor to estimate the mode of operation. A grid filter in Bayesian framework is then applied to the dynamic time warp estimate to provide an accurate posterior estimate of the operational mode of the MFR. This novel approach is tested on an EW scenario via simulation by employing a hypothetical MFR. Based on the simulation results, we conclude that the hybrid echo state network is more suitable than its recurrent multi-layer perceptron counterpart for the mode estimation problem of a RWR.</p> / Thesis / Master of Applied Science (MASc)

Classification of human movement is a large field of interest to Human-Machine Interface researchers. The reason for this lies in the large emphasis humans place on gestures while communicating with each other and while interacting with machines. Such gestures can be digitized in a number of ways, including both passive methods, such as cameras, and active methods, such as wearable sensors. While passive methods might be the ideal, they are not always feasible, especially when dealing in unstructured environments. Instead, wearable sensors have gained interest as a method of gesture classification, especially in the upper limbs. Lower arm movements are made up of a combination of multiple electrical signals known as Motor Unit Action Potentials (MUAPs). These signals can be recorded from surface electrodes placed on the surface of the skin, and used for prosthetic control, sign language recognition, human machine interface, and a myriad of other applications. In order to move a step closer to these goal applications, this thesis compares three different machine learning tools, which include Hidden Markov Models (HMMs), Support Vector Machines (SVMs), and Dynamic Time Warping (DTW), to recognize a number of different gestures classes. It further contrasts the applicability of these tools to noisy data in the form of the Ninapro dataset, a benchmarking tool put forth by a conglomerate of universities. Using this dataset as a basis, this work paves a path for the analysis required to optimize each of the three classifiers. Ultimately, care is taken to compare the three classifiers for their utility against noisy data, and a comparison is made against classification results put forth by other researchers in the field. The outcome of this work is 90+ % recognition of individual gestures from the Ninapro dataset whilst using two of the three distinct classifiers. Comparison against previous works by other researchers shows these results to outperform all other thus far. Through further work with these tools, an end user might control a robotic or prosthetic arm, or translate sign language, or perhaps simply interact with a computer. / Master of Science

Les données manquantes constituent un challenge commun en reconnaissance de forme et traitement de signal. Une grande partie des techniques actuelles de ces domaines ne gère pas l'absence de données et devient inutilisable face à des jeux incomplets. L'absence de données conduit aussi à une perte d'information, des difficultés à interpréter correctement le reste des données présentes et des résultats biaisés notamment avec de larges sous-séquences absentes. Ainsi, ce travail de thèse se focalise sur la complétion de larges séquences manquantes dans les séries monovariées puis multivariées peu ou faiblement corrélées. Un premier axe de travail a été une recherche d'une requête similaire à la fenêtre englobant (avant/après) le trou. Cette approche est basée sur une comparaison de signaux à partir d'un algorithme d'extraction de caractéristiques géométriques (formes) et d'une mesure d'appariement élastique (DTW - Dynamic Time Warping). Un package R CRAN a été développé, DTWBI pour la complétion de série monovariée et DTWUMI pour des séries multidimensionnelles dont les signaux sont non ou faiblement corrélés. Ces deux approches ont été comparées aux approches classiques et récentes de la littérature et ont montré leur faculté de respecter la forme et la dynamique du signal. Concernant les signaux peu ou pas corrélés, un package DTWUMI a aussi été développé. Le second axe a été de construire une similarité floue capable de prender en compte les incertitudes de formes et d'amplitude du signal. Le système FSMUMI proposé est basé sur une combinaison floue de similarités classiques et un ensemble de règles floues. Ces approches ont été appliquées à des données marines et météorologiques dans plusieurs contextes : classification supervisée de cytogrammes phytoplanctoniques, segmentation non supervisée en états environnementaux d'un jeu de 19 capteurs issus d'une station marine MAREL CARNOT en France et la prédiction météorologique de données collectées au Vietnam. / Missing data are a prevalent problem in many domains of pattern recognition and signal processing. Most of the existing techniques in the literature suffer from one major drawback, which is their inability to process incomplete datasets. Missing data produce a loss of information and thus yield inaccurate data interpretation, biased results or unreliable analysis, especially for large missing sub-sequence(s). So, this thesis focuses on dealing with large consecutive missing values in univariate and low/un-correlated multivariate time series. We begin by investigating an imputation method to overcome these issues in univariate time series. This approach is based on the combination of shape-feature extraction algorithm and Dynamic Time Warping method. A new R-package, namely DTWBI, is then developed. In the following work, the DTWBI approach is extended to complete large successive missing data in low/un-correlated multivariate time series (called DTWUMI) and a DTWUMI R-package is also established. The key of these two proposed methods is that using the elastic matching to retrieving similar values in the series before and/or after the missing values. This optimizes as much as possible the dynamics and shape of knowledge data, and while applying the shape-feature extraction algorithm allows to reduce the computing time. Successively, we introduce a new method for filling large successive missing values in low/un-correlated multivariate time series, namely FSMUMI, which enables to manage a high level of uncertainty. In this way, we propose to use a novel fuzzy grades of basic similarity measures and fuzzy logic rules. Finally, we employ the DTWBI to (i) complete the MAREL Carnot dataset and then we perform a detection of rare/extreme events in this database (ii) forecast various meteorological univariate time series collected in Vietnam

Ce manuscrit porte sur la modélisation et l'estimation de certaines non-stationnarités dans les signaux audio. Nous nous intéressons particulièrement à une classe de modèles de sons que nous nommons timbre*dynamique dans lesquels un signal stationnaire, associé au phénomène physique à l'origine du son, est déformé au cours du temps par un opérateur linéaire unitaire, appelé opérateur de déformation, associé à l'évolution temporelle des caractéristiques de ce phénomène physique. Les signaux audio sont modélisés comme des processus gaussiens généralisés et nous donnons dans un premier temps un ensemble d'outils mathématiques qui étendent certaines notions utilisées en traitement du signal au cas des processus stochastiques généralisés.Nous introduisons ensuite les opérateurs de déformations étudiés dans ce manuscrit. L'opérateur de modulation fréquentielle qui est l'opérateur de multiplication par une fonction à valeurs complexes de module unité, et l'opérateur de changement d'horloge qui est la version unitaire de l'opérateur de composition.Lorsque ces opérateurs agissent sur des processus stationnaires les processus déformés possèdent localement des propriétés de stationnarité et les opérateurs de déformation peuvent être approximés par des opérateurs de translation dans les plans temps-fréquence et temps-échelle. Nous donnons alors des bornes pour les erreurs d'approximation correspondantes. Nous développons ensuite un estimateur de maximum de vraisemblance approché des fonctions de dilatation et de modulation. L'algorithme proposé est testé et validé sur des signaux synthétiques et des sons naurels. / This manuscript deals with the modeling and estimation of certain non-stationarities in audio signals. We are particularly interested in a sound class models which we call dynamic*timbre in which a stationary signal, associated with the physical phenomenon causing the sound, is deformed over time by a linear unitary operator, called deformation operator, associated with the temporal evolution of the characteristics of this physical phenomenon.Audio signals are modeled as generalized Gaussian processes. We give first a set of mathematical tools that extend some classical notions used in signal processing in case of generalized stochastic processes.We then introduce the two deformations operators studied in this manuscript. The frequency modulation operator is the multiplication operator by a complex-valued function of unit module and the time-warping operator is the unit version of the composition operator by a bijective function.When these operators act on generalized stationary processes, deformed process are non-stationary generalized process which locally have stationarity properties and deformation operators can be approximated by translation operators in the time-frequency plans and time-scale.We give accurate versions of these approximations, as well as bounds for the corresponding approximation errors.Based on these approximations, we develop an approximated maximum likelihood estimator of the warping and modulation functions. The proposed algorithm is tested and validated on synthetic signals. Its application to natural sounds confirm the validity of the timbre*dynamic model in this context.

The aim of this project is to use NI myRIO and LabVIEW for voice controlled home automation. The NI myRIO is an embedded device which has a Xilinx FPGA and a dual-core ARM Cortex-A9processor as well as analog input/output and digital input/output, and is programmed with theLabVIEW, a graphical programming language. The voice control is implemented in two differentsystems. The first system is based on an Amazon Echo Dot for voice recognition, which is acommercial smart speaker developed by Amazon Lab126. The Echo Dot devices are connectedvia the Internet to the voice-controlled intelligent personal assistant service known as Alexa(developed by Amazon), which is capable of voice interaction, music playback, and controllingsmart devices for home automation. This system in the present thesis project is more focusingon myRIO used for the wireless control of smart home devices, where smart lamps, sensors,speakers and a LCD-display was implemented. The other system is more focusing on myRIO for speech recognition and was built on myRIOwith a microphone connected. The speech recognition was implemented using mel frequencycepstral coefficients and dynamic time warping. A few commands could be recognized, includinga wake word ”Bosse” as well as other four commands for controlling the colors of a smart lamp. The thesis project is shown to be successful, having demonstrated that the implementation ofhome automation using the NI myRIO with two voice-controlled systems can correctly controlhome devices such as smart lamps, sensors, speakers and a LCD-display.

Dynamic programming has been widely used to solve various kinds of optimization problems.In this work, we show that two crucial problems in video-based sign language and gesture recognition systems can be attacked by dynamic programming with additional multiple observations. The first problem occurs at the higher (sentence) level. Movement epenthesis [1] (me), i.e., the necessary but meaningless movement between signs, can result in difficulties in modeling and scalability as the number of signs increases. The second problem occurs at the lower (feature) level. Ambiguity of hand detection and occlusion will propagate errors to the higher level. We construct a novel framework that can handle both of these problems based on a dynamic programming approach. The me has only be modeled explicitly in the past. Our proposed method tries to handle me in a dynamic programming framework where we model the me implicitly. We call this enhanced Level Building (eLB) algorithm. This formulation also allows the incorporation of statistical grammar models such as bigrams and trigrams. Another dynamic programming process that handles the problem of selecting among multiple hand candidates is also included in the feature level. This is different from most of the previous approaches, where a single observation is used. We also propose a grouping process that can generate multiple, overlapping hand candidates. We demonstrate our ideas on three continuous American Sign Language data sets and one hand gesture data set. The ASL data sets include one with a simple background, one with a simple background but with the signer wearing short sleeved clothes, and the last with a complex and changing background. The gesture data set contains color gloved gestures with a complex background. We achieve within 5% performance loss from the automatically chosen me score compared with the manually chosen me score. At the low level, we first over segment each frame to get a list of segments. Then we use a greedy method to group the segments based on different grouping cues. We also show that the performance loss is within 5% when we compare this method with manually selected feature vectors.

Le sujet de la thèse est l'analyse de la variabilité de forme d'un ensemble de signaux avec comme principales applications le traitement des signaux électrophysiologiques mesurés sur la chaîne auditive et le cœur. Cette variabilité de forme des signaux est souvent présente dans les signaux issus de processus naturels et elle est porteuse d'information. Pour accéder à cette information, il est nécessaire de formaliser le concept d'écart de forme et de proposer des outils statistiques spécifiques. Certaines méthodes, issues de la communauté statistique, ont été récemment proposées pour analyser la variabilité présente dans un ensemble de signaux. Ces méthodes travaillent dans un cadre fonctionnel en considérant les données comme des observations de fonctions. Elles cherchent à éliminer la variabilité temporelle dans le but d'accéder à une variabilité d'amplitude par divers algorithmes dit de « recalage de courbes ». Dans cette thèse nous proposons de nouvelles méthodes d'analyse de forme qui utilisent aussi un réalignement temporel (ang : time warping) mais dont le sens diffère des approches de recalage de courbes. De plus, ces méthodes proposent une moyenne de forme et distance de forme permettant la mesure de la variabilité de forme. Au chapitre 1, nous présentons toutes ces méthodes et les comparons afin d'aider l'utilisateur à bien choisir suivant l'application dédiée. <br /><br />Au chapitre 2, nous nous intéressons à la caractérisation objective de l'acouphène, une sensation sonore fantôme. En effet, un problème majeur est l'absence de critère objectif pour le caractériser. Pour cela nous étudions l'activité spontanée composite (ASC) issue du nerf auditif et les potentiels évoqués (PE) issus de relais auditifs en présence de salicylate, un générateur d'acouphènes, chez le cochon d'Inde. La première partie du travail consiste en la présentation d'un modèle de génération de l'ASC. Ce modèle nous sert à tester en simulation des scénarios possibles d'altérations neurosensorielles en présence de salicylate. En complément de l'index spectral décrit dans la littérature, nous proposons d'employer un critère de similarité sur la distribution d'amplitude de l'ASC pour mesurer ces altérations. La seconde partie du chapitre consiste à étudier la variabilité temporelle des PE sur plusieurs relais auditifs en présence de salicylate. <br /><br />Au chapitre 3, nous montrons des applications de détection de pathologies à partir de l'analyse de forme d'une composante spécifique de l'ECG, l'onde P. Les pathologies concernées sont la fibrillation auriculaire et l'apnée du sommeil.

The reseach on neural networks has been done for six decades. In this period, many neural models and learning rules have been proposed. Futhermore, they were popularly and successfully applied to many applications. They successfully solved many problems that traditional algorithms could not solve efficiently . However, applying multilayer neural networks to applications, users are confronted with the problem of determining the number of hidden layers and the number of hidden neurons in each hidden layer. It is too difficult for users to determine proper neural network architectures. However, it is very significant, because neural network architectures always influence critically their performance. We may solve problems efficiently, only when we has proper neural network architectures. To overcome this difficulty, several approaches have been proposed to generate the architecture of neural networks recently. However, they still have some drawbacks. The goal of our research is to discover better approachs to automatically determine proper neural network architectures. We propose a series of approaches in this thesis. First, we propose an approach based on decision trees. It successfully determines neural network architectures and greatly decreases learning time. However, it can deal only with two-class problems and it generates bigger neural network architectures. Next, we propose an information entropy based approach to overcome the above drawbacks. It can generate easily multi-class neural networks for standard domain problems. Finally, we expand the above method for sequential domain and structured domain problems. Therefore, our approaches can be applied to many applications. Currently, we are trying to work on quantum neural networks. We are also interested in ART neural networks. They are also incremental neural models. We apply them to digital signal processing. We propose a character recognition application, a spoken word recognition application, and an image compression application. All of them have good performances.

Cette étude consiste à étudier l'apport de réseaux de neurones<br />artificiels pour améliorer le contrôle de processus industriels<br />complexes, caractérisés en particulier par leur aspect temporel.<br />Les motivations principales pour traiter des séries temporelles<br />sont la réduction du volume de données, l'indexation pour la<br />recherche de similarités, la localisation de séquences,<br />l'extraction de connaissances (data mining) ou encore la<br />prédiction.<br /><br />Le processus industriel choisi est un four à arc<br />électrique pour la production d'acier liquide au Luxembourg. Notre<br />approche est un concept de contrôle prédictif et se base sur des<br />méthodes d'apprentissage non-supervisé dans le but d'une<br />extraction de connaissances.<br /><br />Notre méthode de codage se base sur<br />des formes primitives qui composent les signaux. Ces formes,<br />composant un alphabet de codage, sont extraites par une méthode<br />non-supervisée, les cartes auto-organisatrices de Kohonen (SOM).<br />Une méthode de validation des alphabets de codage accompagne<br />l'approche.<br /><br />Un sujet important abordé durant ces recherches est<br />la similarité de séries temporelles. La méthode proposée est<br />non-supervisée et intègre la capacité de traiter des séquences de<br />tailles variées.

Dynamic speech properties such as time warping, silence removal and background noise interference are the most challenging issues in continuous speech signal matching. Among all of them, the time warped speech signal matching is of great interest and has been a tough challenge for the researchers. An adaptive framing based continuous speech tracking and similarity measurement approach is introduced in this work following a comprehensive research conducted in the diverse areas of speech processing. A dynamic state model is introduced based on system of linear motion equations which models the input (test) speech signal frame as a unidirectional moving object along the template speech signal. The most similar corresponding frame position in the template speech is estimated which is fused with a feature based similarity observation and the noise variances using a Kalman filter. The Kalman filter provides the final estimated frame position in the template speech at current time which is further used for prediction of a new frame size for the next step. In addition, a keyword spotting approach is proposed by introducing wavelet decomposition based dynamic noise filter and combination of beliefs. The Dempster’s theory of belief combination is deployed for the first time in relation to keyword spotting task. Performances for both; speech tracking and keyword spotting approaches are evaluated using the statistical metrics and gold standards for the binary classification. Experimental results proved the superiority of the proposed approaches over the existing methods.