Search

The present paper introduces an extension of the quantile periodogram developed in Li (2012) since the latter suffers from instabilities and spectral leakage. To overcome these drawbacks, a L 1 regularised version is proposed. The asymptotic properties of the new spectral estimator are discussed and simulations are performed in order to show its efficiency in the presence of hidden periodicities and under different types of noise.

International audience; A robust neural adaptive integral sliding mode control approach is proposed in the present paper for nonlinear exoskeleton systems. The proposed control technique is composed of two parts: an adaptive neural network controller and an adaptive integral terminal sliding mode controller. The adaptive laws are developed to estimate unknown parameters and ensure asymptotic stability of the closed-loop system. Only classical system's properties are supposed to be known, such as the bounds on some parameters. The unknown dynamics of the system are estimated on line by the neural network control part. The proposed adaptive control strategy is designed to ensure the reaching of the sliding surface with enhanced tracking performance. The singularity problem of the terminal sliding mode approach is overcomed without adding any constraint. The closed-loop stability of the system in the sense of Lyapunov is demonstrated. The effectiveness of the proposed approach is tested in real time application with healthy human subjects by performing passive arm movements using a 2-DOF upper limb exoskeleton.

International audience; The Internet of Things (IoT) refers to the domain of physical/logical objects connected to the Internet that can be accessible anywhere and anytime, giving the access policies. Web objects and services can be seen as distributed and cooperative agents that need to collaborate in order to reach advanced functionalities and also to optimize the overall quality offered to the end users. However, the number of the services that can be supplied through these devices on the Internet will increase more and more. In this context, performing the composition and selection of the best services according to quality-of-service (QoS) criteria become a complex task, especially when dealing with global service selection to satisfy global constraints of users in decentralized environments. For this purpose, we propose a distributed and optimal QoS selection approach based on the multiagent paradigm and the distributed constraint optimization problem (DCOP) formalism. Hence, we revisited a DCOP technique for developing an efficient algorithm: the DPOP4QoS implemented under a real-time Web protocol for IoT communication, the extensible messaging and presence protocol. The proposed algorithm takes into account the specificities of the service composition context and the satisfaction of the global user constraints. This paper also presents a set of experiments conducted under realistic distributed environment to evaluate the performance of the proposed algorithm.

International audience; This brief deals with a modular controller using a fast terminal sliding mode approach for articulated systems represented by exoskeletons to perform flexion/extension movements. The proposed controller supposes that all model functions are unknown except classical properties related to the boundedness of some parameters. On the other hand, the disturbances are assumed to be bounded. It permits finite-time convergence to the desired trajectories in both position and velocity. The system is divided into several subsystems and a particular unit controls each subsystem. A supervisor using the Lyapunov approach ensures the closed-loop stability of the overall system. The proposed robust controller has been applied in a real-time application to drive an upper limb exoskeleton having 3 DOF. The used device worn by a healthy subject performs flexion/extension movements often practiced for rehabilitation purposes. A strict security protocol, which is generally used by therapists, has been respected. The obtained results are satisfactory and prove the effectiveness and the robustness of the proposed controller.

International audience; Challenges for the next generation of Brain Computer Interfaces (BCI) are to mitigate the common sources of variability (electronic, electrical, biological) and to develop online and adaptive systems following the evolution of the subject's brain waves. Studying electroencephalographic (EEG) signals from their associated covariance matrices allows the construction of a representation which is invariant to extrinsic perturbations. As covariance matrices should be estimated, this paper first presents a thorough study of all estimators conducted on real EEG recording. Working in Euclidean space with covariance matrices is known to be error-prone, one might take advantage of algorithmic advances in Riemannian geometry and matrix manifold to implement methods for Symmetric Positive-Definite (SPD) matrices. Nonetheless, existing classification algorithms in Riemannian spaces are designed for offline analysis. We propose a novel algorithm for online and asynchronous processing of brain signals, borrowing principles from semi-unsupervised approaches and following a dynamic stopping scheme to provide a prediction as soon as possible. The assessment is conducted on real EEG recording: this is the first study on Steady-State Visually Evoked Potential (SSVEP) experimentations to exploit online classification based on Rie-mannian geometry. The proposed online algorithm is evaluated and compared with state-of-the-art SSVEP methods, which are based on Canonical Correlation Analysis (CCA). It is shown to improve both the classification accuracy and the information transfer rate in the online and asynchronous setup.

International audience; Brain Computer Interfaces (BCI) based on electroencephalog-raphy (EEG) rely on multichannel brain signal processing. Most of the state-of-the-art approaches deal with covariance matrices , and indeed Riemannian geometry has provided a substantial framework for developing new algorithms. Most notably , a straightforward algorithm such as Minimum Distance to Mean yields competitive results when applied with a Riemannian distance. This applicative contribution aims at assessing the impact of several distances on real EEG dataset , as the invariances embedded in those distances have an influence on the classification accuracy . Euclidean and Riemannian distances and means are compared both in term of quality of results and of computational load .

International audience; This study deals with the bounded control of a lower limb exoskeleton having four (4) Degree of Freedom (DoF), which are located at hips and knees joint level. It is intended to be used as a rehabilitative device for individuals with hip or knee joint impairments. The rehabilitation protocol entails using Central Pattern Generators (CPGs) to design rhythmic movement close to that of the human gait. In order to remedy the drawback associated with CPGs which is characterised by the inability to choose specified initial conditions for the generated trajectories, the use of time polynomial functions is employed to approximate the joint angle profile generated. This gait is meant to be tracked by the user under the influence of a controller. The control torque is envisaged to ensure a proper trajectory following of the exoskeleton for the flexion/extension movements of the hip and knee within the range of motion associated with walking. With this approach, the wearer is completely passive. Using a bounded controller further protect the user and the device by ensuring the actuator's physical limits are not exceeded. Simulation results with the aid of Matlab/Simulink does help validate the proposed assertion.

Graphical abstractDisplay Omitted HighlightsHuman safety is ensured for the wearer of exoskeleton.Adaptive controller without prior knowledge.Stability proved theoretically using Lyapunov approach.No chattering phenomena in the input control.External disturbances are rejected and the wearer efforts are taken into account. This paper deals with the rehabilitation purposes using an active orthosis driven by an adaptive neural controller based on a radial basis function neural network (RBFNN). Two essential conditions are required in our study: ensuring the wearer safety and the good trajectory tracking. We consider for our experiments the same movements often recommended by the doctor during therapy sessions. In this context, it is possible to add some trivial prior knowledge as the dynamic model structure and all dynamical identified parts. The unknown or the uncertainty part of the inertia term of the knee-shank-orthosis system is identified online using an adaptive term. All other uncertainties or unknown dynamics are identified online by the RBFNN. The Lyapunov approach has been used to derive adaptation laws of the neural parameters and the inertia term. These adaptation laws ensure the stability of the system composed of the exoskeleton and its wearer. The wearer can be completely inactive or applying either a resistive or an assistive effort. Experimental results have been conducted on a real exoskeleton that is used for rehabilitation reasons. Based on these results we conclude with the effectiveness of the proposed approach.

International audience; In this paper we propose an adaptive controller based on sliding mode and backstepping approaches. The system to be controlled is an exoskeleton used for the rehabilitation of the knee joint. The person wearing the exoskeleton is healthy, sitting and performs movements of flexion/extension. This kind of movement is usually applied by the doctor. The parameters of the dynamic model of the overall system (knee & exoskeleton) are considered unknown. The proposed controller is stable and robust against bounded external disturbances. The experimental results show the effectiveness of the proposed approach. The tests have also been validated when the wearer applies a resistive/assistive torque.

International audience; In this paper, an inverse dynamics approach by means of adaptive coupled oscillators is used in the modelling and control of a lower limb orthosis applied at the knee and ankle joint level. This design is aimed at providing assistance and rehabilitative measures to humans with lower limb disorders and as such presents a platform for which their mobility performance can be improved. Adaptive oscillators are known to have the capability of learning high level parameters of sinusoidal, quasi-sinusoidal or non-sinusoidal signals (amplitude, frequency and offset). However, the later signal (non-sinusoidal) considered in this paper requires a number of oscillators in parallel to replicate the moving joint regarding filtering via adaptive oscillator. The dynamic model for the knee and ankle are considered to take the form of a damped pendulum model connected by two revolute joints. This maps the input torque of both joints (knee and ankle) to their output trajectories, hence integrating the different forces at the joint level of the different joints. The coupling effect is achieved by the use of coupled Adaptive Frequency Oscillator (AFO) for the estimation of the joint trajectories. Tracking performance for the knee-ankle orthosis is studied for non-sinusoidal reference trajectories, having a global coupling between the joints. The results obtained using SCILAB show a good performance of the controller trajectory tracking capabilities even in the presence of external disturbances.

International audience; In this paper we propose to drive an actuated orthosis using an adaptive controller based on a reference model. It is not necessary to know all the functions of the dynamic model. Needing only the global structure of the dynamic model, we use a specific adaptive controller to obtain good performance in terms of trajectory tracking both in position and in velocity. A Multi-Layer Perceptron Neural Network (MLPNN) is used to estimate dynamics related to inertia, gravitational and frictional forces along with other unmodeled dynamics. The Lyapunov formalism is used for stability study of the system (shank+orthosis) in closed loop and to determine adaptation laws of the neural parameters. To treat the non-linearties related to the MLPNN, we have used first order Taylor series expansion. Experimental results have been obtained using a real orthosis worn by an appropriate dummy. Several tests have been realized to verify the effectiveness and the robustness of the proposed controller. For instance, our proposed orthosis model has given robust tracking performance under assistive as well as resistive forces.

International audience; This paper deals with the controller of an active orthosis for rehabilitation reasons of the knee joint. The dynamical model of the system, constituted of the shank and the orthosis is complex and is considered as unknown in the conception of the proposed controller. The full security protocol has been carefully applied and we have selected a healthy person for our experiments. The flexion/extension movements used for our experiments are of sinusoidal form and are generally applied by therapeutic doctors. The fast terminal sliding mode technique used in the proposed controller permits a finite time convergence towards zero of the tracking errors both in position and in velocity. The experimental results are satisfactory and prove clearly the effectiveness of the proposed approach. As the wearer used can develop a muscular effort, we have tested the two cases: resistive and assistive effort and we have obtained a good performance in both cases as has been proven in the stability analysis by the the Lyapunov approach.

International audience; Brain Computer Interfaces (BCI) rely on brain waves signal, such as electro-encephalogram (EEG) recording, to endow a disabled user with non-muscular communication. Given the very low signal-to-noise ratio of EEG, a signal enhancement phase is crucial for ensuring decent performances in BCI systems. Several methods have been proposed for EEG signal enhancement, such as Independent Component Analysis, Common Spatial Pattern, and Principal Component Analysis. We show that Canonical Correlation Analysis (CCA), initially introduced to SSVEP-based BCI as a feature extraction method, is a good candidate for such preprocessing state. Evaluation is performed on a recording from 5 subjects during a BCI task based on Steady-State Visual Evoked Potentials (SSVEP). We demonstrate that CCA significantly improves classification performances in SSVEP-based BCIs.

VANETs are becoming more and more popular as a way to increase the traffic safety and comfort. The IEEE 802.11p standard, especially the 802.11p MAC protocol, has attracted much attention as part of the WAVE protocol in VANETs. Safety applications, as one of the main categories of applications in VANETs, is very challenging for the design of a MAC protocol due to their low latency and high reliability requirements. The CCH interval is also a key parameter for the 802.11p MAC protocol since it can affect the performance of safety message delivery significantly. In this paper, a simulation based evaluation is proposed to evaluate the performance of the 802.11p MAC protocol with various vehicle densities and CCH interval settings. The evaluation results indicate that the 802.11p MAC protocol can be improved via extending the CCH interval. However, the reliability is still very challenging due to high collision rates.

Widespread deployment of wireless local area networks and a gradual increase in streaming applications have brought about a demand for improved quality of service (QoS) in wireless networks. However, increasing user datagram protocol based high priority multimedia traffic and the class differentiation introduced in QoS protocols, has resulted into transmission control protocol (TCP) starvation and increased spurious timeouts. While today's Internet traffic is still dominated by TCP based applications, the negative effects of IEEE 802.11e enhanced distributed coordination function (EDCF) scheme on TCP performance in the presence of high priority traffic have not been extensively explored. In this paper, the performance of TCP in 802.11e WLAN competing with high priority traffic is examined. The prioritised adaptive enhanced scheme (PAD_EDCF) is proposed. The proposed scheme gives priority to TCP control packets in order to improve the low traffic transmission flow and acquires additional capability of adjusting the MAC parameters based on the traffic load condition. Simulation results demonstrate that the proposed scheme significantly improves TCP performances in terms of traffic efficiency, throughput and reduces delay.

International audience; To reduce the risk of accidents, traffic, and other safety related problem's on public roads, the Wireless Access in Vehicular Environment (WAVE) standard was created. The WAVE also known as Dedicated Short Range Communications (DSRC) at 5.9 GHz, is part of the Federal Highway Authority's Vehicle Infrastructure Integration (VII). This standard merely supports Vehicle-to-Vehicle (V2 V) and Vehicle-to-Infrastructure (V2I) communications for emerging Intelligent Transportation Systems (ITS). Due to the high mobility of this network, nodes involved in communication suffer from intermittent signal degradation. This is partly due to Doppler Effect (DE). This paper investigates and analyses the DE over wide Doppler Shift (DS) ranges. The results of the analysis clearly demonstrate that, sustainable communication links is achievable with DS ranging up to 1400 Hz if an appropriate Modulation Code Scheme (MCS) can be selected. It also demonstrates that a BPSK Rate of 1/2 is not always a good candidate when combating DE.

International audience; The wormhole attack is one of the most popular and serious attacks in Wireless sensor networks and most proposed protocols to defend against this attack use extra hardware which impacts highly on the cost of implementation as well causing extra overheads which have high implications on the sensors power consumption. Due to the limited resources in the sensor nodes, protocols developed for wireless sensor networks should not impact heavily on the computational overheads and power consumption in order to extend the network lifetime. In this paper, we exploit the Hidden Markov Model (HMM) Viterbi algorithm, to detect the wormhole attack based on the maximum probabilities computed for a hidden state transition. We use different shortest paths hop count costs between a source and a destination node as the states input to the Viterbi algorithm, earmarking the least cost paths as the suspect wormhole paths, for a given observation sequence of the given shortest paths.

In this paper, we evaluate and analyze the impact of different network loads and varying no. of nodes on distance vector and link state routing algorithms. We select three well known proactive protocols; Destination Sequenced Distance Vector (DSDV) operates on distance vector routing, while Fisheye State Routing (FSR) and Optimized Link State Routing (OLSR) protocols are based on link state routing. Further, we evaluate and compare the effects on the performance of protocols by changing the routing strategies of routing algorithms. We also enhance selected protocols to achieve high performance. We take throughput, End-to-End Delay (E2ED) and Normalized Routing Load (NRL) as performance metrics for evaluation and comparison of chosen protocols both with default and enhanced versions. Based upon extensive simulations in NS-2, we compare and discuss performance trade-offs of the protocols, i.e., how a protocol achieves high packet delivery by paying some cost in the form of increased E2ED and/or routing overhead. FSR due to scope routing technique performs well in high data rates, while, OLSR is more scalable in denser networks due to limited retransmissions through Multi-Point Relays (MPRs)., 48th ICC-AHSN, 2012; Ottawa Canada

The paper presents a new graph-based implementation of bilateral filtering. Based on the Laplacian mesh smoothing framework, the proposed filter mimics the behaviour of the classical mesh filter while retaining some of the interesting properties of mesh smoothing. The comparison between the filters is twofold. First of all, the two filters are benchmarked according to their ability to denoise complex synthetic image transitions. The respective performance of the filters are then assessed in a multiresolution denoising scheme for grayscale images, combining wavelet decomposition, shrinkage and bilateral filtering. The results obtained are encouraging and shows that the BMF is a viable alternative to classical bilateral filtering.

Energy efficiency is a key design criterion for routing protocols in wireless sensor networks since sensor nodes are strongly constrained in terms of energy supply. Gradient-Based Routing (GBR) is a well known energy efficient routing protocol that is used in WSNs. However, there exist shortcomings in the GBR scheme such as: (1) sinks make use of flooding to broadcast interest messages which leads to a lot of duplication packets which are transmitted. This leads to the waste of a lot of energy in the network and (2) nodes deliver messages in a point to point manner. As a result, the potential of data retransmissions in the network is high due to the unstable network environment in WSNs. In this study, network coding and a competitive approach are proposed to solve the above two problems. Firstly, an energy efficient broadcast algorithm using network coding for GBR (GBR-NC) is proposed. This algorithm aims to reduce network traffic, and furthermore, reduce the energy consumption and prolong the lifetime of the network. Secondly, two competing algorithms (GBR-C and auto-adaptable GBR-C) are proposed for GBR. The basic idea of the proposed competing algorithms is to reduce the retransmission attempts and save the energy by considering two forward candidates. Simulation results show that the proposed schemes give better results when compared to the traditional GBR in terms of energy efficiency.

In this paper we present a fault detection strategy for wireless sensor networks. The strategy is based on modeling a sensor node by Takagi-Sugeno-Kang (TSK) fuzzy inference system (FIS), where a sensor measurement of a node is approximated by a function of the sensor measurements of the neighboring nodes. We also model a node by recurrent TSK-FIS (RFIS), where the sensor measurement of the node is approximated as function of real measurements of the neighboring nodes and the previously approximated value of the node itself. Temporary errors in sensor measurements and/or communication are overcome by redundancy of data gathering. A node which has developed a faulty sensor is not completely discarded because it is useful for relaying the information among the other nodes. Each node has its own fuzzy model that is trained with input of neighboring sensors' measurements and an output of its actual measurement. A sensor is declared faulty if the difference between the outcome of the fuzzy model and the actual sensor measurement is greater than the prescribed amount depending on the physical quantity being measured. Simulations are performed using the fuzzy logic toolbox of Matlab. We also give a comparison of obtained results to those from a feed-forward artificial neural network, recurrent neural network and the median [1] of measured values of the neighboring nodes.

Parallel manipulators have been proposed to overcome accuracy problem in the end effector positioning, exhibited by serial manipulators(Stewart, 1965)(Reboulet, 1988)(Merlet, 2000). These parallel robots are primarly used in the applications for which the considered processes require a high degree of accuracy, high speeds or accelerations. Aircraft simulator (Stewart, 1965), machining tools (Neugebauer et al., 1998)(Poignet et al., 2002), and various other medical applications (Merlet, 2002)(Leroy et al., 2003)(Plitea et al., 2008) constitute some of the many possible applications of parallel robots. The computation of the inverse dynamic model is essential for an effective robot control. In the field of parallel robots, many approaches have been developed for efficient computation of the inverse dynamics. The formalism of d’Alembert has been used to obtain an analytical expression of the dynamics model (Fichter, 1986)(Nakamura & Ghodoussi, 1989). The principle of virtual works has been applied in (Tsai, 2000) for solving the inverse dynamics of the Gough-Stewart platform and in (Zhu et al., 2005) for a Tau parallel robot. Lagrangian formalism is applied in (Leroy et al., 2003) for the dynamics modeling of a parallel robot used as a haptic interface for a surgical simulator. These various approaches do not seem effective for a robot dynamic control under the real time constraint. A better computational efficiency can be achieved by the development of approaches using recursive schemes, in particular, based on the Newton-Euler formulation. Gosselin (Gosselin, 1996) proposed an approach for the computation of the inverse dynamic model of planar and spatial parallel robots, in which all the masses and inertias are taken into account. This proposed method is difficult to generalize for all the parallel architectures. Dasgupda et al (Dasgupta & Choudhury, 1999) applied this method to several parallel manipulators. Khan (Khan, 2005) has developed a recursive algorithm for the inverse dynamics. This method is applied to a 3R planar parallel robot. Bi et al (Bi & Lang, 2006) use the Newton-Euler recursive scheme for the computation of the articular forces of a tripod system. Khalil et al (Khalil & Guegan, 2004) proposed a general method for the inverse and direct dynamic model computation of parallel robots, which is applied to several parallel manipulators (Khalil & Ibrahim, 2007). Despite the large amount of contributions in this field, there is still a need for improving the computational effeciency of the inverse kinematic and dynamic model clculation for real-time control. In this paper, a parallel robot is considered as a multi robot system with 17

International audience; The classification of subscriber types in mobile cellular networks is valuable for network service providers since it provides a mechanism to plan network services by better understanding subscriber behaviour in a network. Mobile networks contain vast repositories of data that store valuable information regarding subscriber behaviour. In this paper, a new approach for subscriber classification in mobile cellular networks is proposed. The proposed approach considers network traffic generated from a mobile cellular network operator in South Africa. The proposed approach makes use of a difference histogram approach for feature extraction and a fuzzy c-means clustering algorithm to classify traffic data into subscriber classes. To validate the proposed approach, a comparative analysis of two different multi-resolution feature extraction approaches, the empirical mode decomposition (EMD) approach and the discrete wavelet packet transform (DWPT) are compared with results obtained with the difference histogram (DH) approach. It is shown that the difference histogram provides better clustering results when compared to the two multi-resolution approaches demonstrating the potential of the difference histogram approach.

To optimize performance of ad hoc networks, the contradicting objectives: energy conservation and high data rate realization need to be jointly considered. In such a case, network interference need to be controlled in a manner that users transmit at minimum power enough to sustain the transmission. This paper proposes a joint power and rate adaptation in ad hoc network based on coupled interference minimization (PRIM) where the formulated network utility maximization (NUM) is solved using reverse engineering based on Karush-Kuhn-Tucker (KKT) conditions. In that way, users determine their transmit power and data rate based on their local observations (coupled interference). Pricing mechanism is employed in PRIM to restrict users from self-interest behaviours. It is demonstrated theoretically that PRIM satisfies the conditions of the super-modular games such that the local optimality is always the global optimality. Simulation results have shown that adapting transmits power and data rates based on the coupled interference (i.e. link dynamics) can improve the performance of ad hoc networks.

The discretization of a HT space results in the ρ value detection error of a line. This paper addresses the ρ-direction precision improvement by compensating for this error. The mage are vertically or horizontally shifted, and a series peak positions are obtained by applying the standard HT (SHT) on these shifted images. The change of ρ value of a line due to these shifts is studied. On one hand this change can be measured by detecting the peak shift in HT space, on the other hand it can be obtained by geometric analysis on the vertical or horizontal shift in image space. The difference between the unit change in HT space (i.e. the ρ-direction resolution Δρ) and the unit change in image space (i.e. ρ value change due to vertical or horizontal unit shift) is used to measure the SHT ρ value detection error. A high precision ρ value is obtained by compensating for this error. The experiments show the effectiveness of the proposed method.

In order to guarantee the QoS requirements in IEEE 802.11 WLANs, the IEEE 802.11e standard for MAC enhancement has been proposed to support delay-sensitive applications. The IEEE 802.11e enhanced distributed coordination function (EDCF) provides traffic differentiation based on a set of static Access Category (AC) MAC parameters which guarantee prioritization of real-time flows with strict bandwidth and delay requirements. This reduces the network capacity available for best-effort flows in heterogeneous traffic environments. While today's Internet traffic is still dominated by TCP based applications, the negative effects of IEEE 802.11e service differentiation scheme on TCP performance have not been extensively explored. In this paper, the performance of TCP in 802.11e WLAN competing with high priority traffic is evaluated. A simple TCP-based scheme, called Prioritized Adaptive EDCF, is proposed to improve the existing IEEE 802.11e EDCF mechanism by assigning the highest priority to TCP control packets and adjusting MAC parameters accordingly. The results demonstrate that the proposed scheme improves TCP performance in terms of traffic efficiency, throughput and delay.

International audience; In this paper, we identify and analyze the requirements to design a new routing link metric for wireless multihop networks. Considering these requirements, when a link metric is proposed, then both the design and implementation of the link metric with a routing protocol become easy. Secondly, the underlying network issues can easily be tackled. Thirdly, an appreciable performance of the network is guaranteed. Along with the existing implementation of three link metrics Expected Transmission Count (ETX), Minimum Delay (MD), and Minimum Loss (ML), we implement inverse ETX; invETX with Optimized Link State Routing (OLSR) using NS-2.34. The simulation results show that how the computational burden of a metric degrades the performance of the respective protocol and how a metric has to trade-off between different performance parameters.

In order to take into account the uncertainties about the antecedent membership functions and consequent parameter values in the type-1 TSK Fuzzy Logic System (FLS), this latter has been extended to its type-2 counterpart. The Type-2 TSK FLS contains three different architectures. One of these architectures which handles the uncertainties of the consequent parameter values, when we are interested in the defuzzified output, gives the same result as an equivalent Type-1 TSK FLS. In order to keep the type-2 nature of this architecture and to reduce the computation complexity at its design stage, we propose a new design approach in which the consequent parameters are tuned using multiple output training data that correspond to the same inputs. To show the effectiveness and the feasibility of the proposed approach, we use a time series prediction example as a use case.

The present paper focuses on sharpness enhancement and noise removal in two dimensional gray scale images. In the grid smoothing approach, the image is represented by a graph in which the nodes represent the pixels and the edges reflect the connectivity. A cost function is defined using the spatial coordinates of the nodes and the gray levels present in the image. The minimisation of the cost function leads to new spatial coordinates for each node. Using an adequate cost function, the grid is compressed in the regions with large gradient values and relaxed in the other regions. The result is a grid which fits accurately the objects in the image. In the presented framework, the noise in the initial image is removed using a mesh smoothing approach. The edges are then enhanced using the grid smoothing. If the level of noise is low, the grid smoothing is applied directly to the image. The mathematical framework of the method is introduced in the paper. The processing chain is tested on natural images.

This paper describes an algorithm for a visual human-machine interface that infers a person's intention from the motion of the hand. Work in progress shows a proof of concept tested on static images. The context for which this solution is intended is that of wheelchair bound individuals whose intentions are the direction and speed variation of the wheelchair. Results show that the symmetry property of the hand in motion can serve as an intent indicator.

Multi-Radio Multi-Channel (MRMC) systems are key to power control problems in WMNs. Previous studies have emphasized throughput maximization in such systems as the main design challenge and transmission power control treated as a secondary issue. In this paper, we present an autonomous power adaptation for MRMC WMNs. The transmit power is dynamically adapted by each network interface card (NIC) in response to the locally available energy in a node, queue load, and interference states of a channel. To achieve this, WMN is first represented as a set of Unified Channel Graphs (UCGs). Second, each NIC of a node is tuned to a UCG. Third, a power selection MRMC unification protocol (PMMUP) that coordinates Interaction variables (IV) from different UCGs and Unification variables (UV) from higher layers is proposed. PMMUP coordinates autonomous power optimization by the NICs of a node. The efficacy of the proposed method is investigated through simulations.

Multi-Radio Multi-Channel (MRMC) systems are key to power control problems in WMNs. In this paper, we present a dynamic power control for MRMC WMNs. First, WMN is represented as a set of disjoint Unified Channel Graphs (UCGs). Second, each radio assigned to a unique UCG adjusts the transmission power locally using predicted interference states among different adjacent UCGs. A new power selection MRMC unification protocol (PMMUP) is proposed that coordinates local power optimizations at the radios of a node. The throughput and energy performance of the proposed method is investigated through simulations.

Improving Fuzzy Logic System (FLS) design is of main interest. Linguistic rules of a FLS can be converted into Fuzzy Basis Functions (FBFs). Moreover, numerical rules and their FBFs can be extracted from numerical training data. This combination of both linguistic and numerical information simultaneously makes the FBFs very useful. Since a specific FLS can be expressed as a linear combination of FBFs, we use generalized orthogonality principle on FBFs, that results in a better FLS. In this work, we consider each of these FBFs as a basis vector. We compute the optimal parameters of consequents that make the error vectors orthogonals to these FBFs, resulting in minimization of the magnitudes of these error vectors and consequently of the optimization of the FLS. This design method is used to tune the consequent parameters of a Fuzzy Logic Controller (FLC) for the non linear inverted pendulum on a cart. Simulation results show that a better control performance is achieved.

The Narrowband-Internet of Things (NB-IoT) is one of the leading licensed Low Power Wide Area Networks (LPWANs) that is expected to play a very important role in connecting the exponentially increasing number of connected devices that make up the Internet of Things (IoT). To enable its network scalability, NB-IoT requires more efficient utilization of its limited energy, channel capacity, and spectrum network resources. These characteristics place various constraints in terms of the design of the physical (PHY) channel coding and spectrum access as well as the network layer protocol for the NB-IoT. A few existing channel coding approaches have been proposed to enhance the NB-IoT’s network reliability, and therefore, enable long-distance transmissions. However, most of the existing NB-IoT channel coding approaches are based on repeating transmission data and control signals and do not consider the channel conditions. Although these repetitions ensure network reliability, they are energy expensive and compromise the performance of the data rate of the network. At the same time, the increasing number of devices to be handled by a single Base Station (BS) within a cell increases the probability for interlink interference which compromises the uplink communication reliability, further increases the number of required data transmission repetitions, and therefore, reduces the overall energy efficiency of the network. To mitigate these identified problems, a multi-layer approach is proposed in this study. Firstly, a channel-aware two-dimensional Energy-Efficient Adaptive Channel Coding (EEACC) approach is considered at the PHY layer. The EEACC not only selects an appropriate modulation coding scheme (MCS) based on the estimated channel conditions (dynamically classified as bad, medium or good based on a periodically assessed BLER performance outcome) but also minimizes the transmission repetition number under a pre-assessed probability of successful transmission. Secondly, at the PHY and network layers, the study considers a two-layers NB-IoT mixed spread spectrum and clustering approach with network coding (N-MANC). At one layer, the approach uses an Adaptive Frequency Hopping Spread Spectrum (AFHSS) technique that uses a lightweight and secure pseudo-random sequence with a varying period T depending on the number of NB-IoT nodes in the considered cell to exploit the channel diversity as a way to mitigate inter-link and cross-technology interference. At the second layer, the mixed approach consists of a cooperative clustering approach based on a defined energy-signal strength gradient used for Cluster Head (CHs) selection, and a network coding (data aggregation) to further boost the overall energy efficiency of the network. A systematic mathematical model of the NB-IoT PHY layer is first developed and modelled. Then, the proposed algorithms are implemented and validated in MATLAB. The obtained simulation results demonstrate that the important QoS metrics such as energy efficiency, network reliability, data rate, transmission latency, network scalability, and lifetime are improved by the proposed algorithms when compared to existing approaches., L'Internet des objets à bande étroite (NB-IoT) est l'un des principaux réseaux étendus à basse consommation énergétique (LPWAN) dans la bande des fréquences sous licence d’utilisation, qui devrait jouer un rôle très important dans la connexion du nombre exponentiellement croissant d'objets connectés qui composent l'Internet des objets (IoT). Pour permettre l'évolutivité de son réseau, la NB-IoT nécessite une utilisation plus efficace de son énergie limitée, de la capacité de ses canaux et des ressources du spectre du réseau. Ces caractéristiques créent ainsi diverses contraintes en termes de conception du codage physique des canaux (PHY) et de l'accès au spectre ainsi que du protocole de la couche réseau pour l'Internet des objets. Quelques approches de codage de canal existantes ont été proposées pour améliorer la fiabilité du réseau de la NB-IoT et permettre ainsi les transmissions à longue distance. Cependant, la plupart des approches de codage de canal existantes de la NB-IoT sont basées sur la répétition des données de transmission et des signaux de contrôle et ne tiennent pas compte des conditions du canal. Bien que ces répétitions assurent la fiabilité du réseau, elles sont coûteuses en énergie et compromettent les performances du réseau en matière de débit de données. En même temps, le nombre croissant de dispositifs devant être traités par une seule station de base (BS) au sein d'une cellule augmente la probabilité d'interférences entre les liaisons, ce qui compromet la fiabilité de la communication sur la liaison montante, augmente encore le nombre requis de répétitions de transmission de données et, par conséquent, réduit l'efficacité énergétique globale du réseau. Pour pallier ces divers problèmes identifiés, une approche multicouche est proposée dans cette thèse. Tout d'abord, une approche bidimensionnelle de codage de canal adaptatif économe en énergie (EEACC) est proposée au niveau de la couche PHY. L'EEACC ne se contente pas de sélectionner un schéma de codage de modulation (MCS) approprié en fonction des conditions estimées du canal (classé dynamiquement comme mauvais, moyen ou bon sur la base d'un résultat de performance BLER évalué périodiquement), mais il minimise également le nombre de répétitions de transmission en fonction d'une probabilité préévaluée de réussite de la transmission. Deuxièmement, en ce qui concerne les couches PHY et réseau, l'étude développe une approche à deux couches NB-IoT `a étalement mixte du spectre et à regroupement avec codage du réseau (N-MANC). Sur une couche, l'approche utilise une technique d'étalement du spectre par saut de fréquence adaptative (AFHSS) qui utilise une séquence pseudo-aléatoire légère et sûre avec une période variable T selon le nombre de nœuds NB-IoT dans la cellule considérée pour exploiter la diversité des canaux comme moyen d'atténuer les interférences entre les liaisons et les technologies croisées. A la deuxième couche, l'approche mixte consiste en une approche de regroupement coopérative basée sur un gradient de force du signal énergétique défini utilisé pour la sélection des têtes de grappe (CHs), et un codage du réseau (agrégation de données) pour stimuler davantage l'efficacité énergétique globale du réseau. Un modèle mathématique systématique de la couche PHY du NB-IoT est d'abord développé et modélisé. Ensuite, les algorithmes proposés sont mis en œuvre et validés utilisant les simulations MATLAB. Les résultats de simulation obtenus montrent que les algorithmes proposés améliorent les paramètres importants de la qualité de service (QoS), tels que l'efficacité énergétique, la fiabilité du réseau, le débit de données, la latence de transmission, l'extensibilité du réseau et la durée de vie, par rapport aux approches existantes.

The quantile periodogram developed by Li(2012) is a new approach that provides an extended and richer breadth of information compared to the ordinary periodogram. However, it suffers from unstable performances where multiple peak frequencies are present because the spectral leak produces extra small spikes. To alleviate this issue, a regularised version of the quantile periodogram is proposed. The asymptotic properties of the new spectral estimator are developed and extensive simulations were performed to show the effectiveness of the proposed estimation in terms of detection of hidden periodicities under different types of noise.A first application of the proposed approach was conducted in a framework of EEG (electroencephalogram) signal study. EEG signals are known for their non-stationarity and non-linearity. The newly proposed estimator was used along with other spectral estimators as feature extraction methods. These features are subsequently fed to classifiers to determine if the signal belongs to one of the motor imagery classes.The second application is a work that has been done in the framework of the European MEDOLUTION project for the study of accelerometer signals. The regularised quantile periodogram as well as other spectral estimators were applied for the classification of self-rehabilitation movements.The results suggest that the regularised quantile periodogram is a promising and robust method for detecting hidden periodicities, especially in cases of non-stationarity and non-linearity of the data, as well as for improving classification performance in the context of EEG and accelerometer signals., Le quantile périodogramme développé par Li (2012) est une nouvelle approche qui fournit une information plus étendue et plus riche par rapport au périodogramme ordinaire. Cependant, il souffre d'instabilité de performance dans le cas de présence de plusieurs périodicités dominantes ou dans le cas de données bruitées. Cela est du à la fuite spectrale (Leakage) produite sous forme de piques supplémentaires. Afin de remédier à ce problème, une version régularisée du quantile périodogramme est proposée. Les propriétés asymptotiques du nouvel estimateur sont développées. De plus, des simulations approfondies sont effectuées pour montrer l’efficacité de l’estimation proposée en termes de détection de périodicités cachées sous différents types de bruit.Une première application de l'approche proposée a été conduite dans un cadre d'étude des signaux EEG (électroencéphalogramme). Les signaux EEG sont connus pour leurs non-stationnarité et non-linéarité. Le nouvel estimateur proposé a été utilisé ainsi que d'autre estimateurs spectraux comme étant des méthodes d'extraction de caractéristiques. Ces caractéristiques sont introduites par la suite à des classifieurs pour déterminer si le signal appartient à une des classes d'imagerie motrice.La deuxième application est un travail qui a été fait dans le cadre du projet Européen MEDOLUTION pour l'étude des signaux accéléromètres. Le quantile périodogramme régularisé ainsi que d'autre estimateurs spectraux ont été appliqués pour la classification des mouvements d'auto rééducation.Les résultats suggèrent que le quantile périodogramme régularisé est une méthode prometteuse et robuste pour détecter des périodicités cachées, spécialement dans les cas de non-stationnarité et non-linéarité des données, ainsi que pour améliorer les performances de classification dans le cadre des signaux EEG et des signaux accéléromètres.

The society is more and more concerned with the aging population and the people with disabilities and the people suffering from injuries. Mobility assistance will reduce dependency and improve the person mobility. For the people suffering mobility problems, some research and studies showed than assistive rehabilitation could contribute to reduce significantly the dependency. The medical and therapeutic services are already operating near full capacity. Therefore, new solutions are needed to improve patient care and attenuate the workload on the medical staff. The robotic systems are considered as part of the healthcare assistive technologies, particularly in rehabilitation phases.Robotic assistive technologies are used to automate part of the therapeutics rehabilitation and give access to new robotics tools to medical staff. Robots have the advantage of their usage for repetitive tasks and exercises. Also, the presence of sensors allows collecting data to evaluate the progress made by the patient during the rehabilitation phases.Exoskeletons are wearable robots based on the structure of the human body to operate on the side. Exoskeletons are made to have the axes of the exoskeleton joints aligned with the human’s joints to be able to provide the needed assistance. The present thesis addresses the problem of the control of a right upper limb exoskeleton for the realisation of movement within the rehabilitation context. The overall control problem is addressed with the human in the control loop, using user’s intention detection based on the EEG signals provided by a Brain Computer Interface (BCI). The research work has been done within the LISSI laboratory (Laboratory Images, Signals and Intelligent Systems) of the UPEC university in France.The first contribution consists of the design of an adaptive control law. It is based on a new switching law, which allows to prevent singularities in the control law and ensuring the convergence of the error to zero. A second proposed control approach of the exoskeleton is based on the estimation by neural networks of the variable dynamic parameters, which are dependent on the morphology of the user. To guarantee the stability of the control strategy a sliding mode controller is added. Indeed, this controller is demonstrated to be robust and ensure the stability of the system regardless the external disturbances and the approximation errors of the neural networks. Finally, the last contribution of this thesis concerns the integration of an EEG (electroencephalography) signal processing part for user’s intention detection., Les progrès scientifiques et particulièrement ceux de la médecine ont permis l'augmentation de l'espérance de vie ainsi que la diminution de la mortalité de certaines pathologies dans les sociétés occidentales. Ces progrès amènent aussi de nouveaux défis: ils ont entrainé par la même occasion un vieillissement de la population et l'augmentation du nombre de personnes en situation de handicap. Ainsi la gestion de la dépendance et de la mobilité est devenue l'un des grands défis des sociétés occidentales. Des études ont montré qu'une bonne rééducation peut contribuer à la diminution de manière significative de cette dépendance. Les services médicaux et thérapeutique étant déjà fortement sollicités, il était donc nécessaire d'apporter de nouvelles solutions technologiques pour améliorer le soin au patient tout en diminuant la charge de travail du personnel médicale. La robotique semble pouvoir apporter des solutions technologiques pour apporter de l'aide au système médical actuel, notamment dans les phases de rééducation. Il serait envisageable d'automatiser une partie de la thérapie de rééducation, et donner accès à de nouveaux outils robotisés d’assistances pour les thérapeutes. Les robots comportent l'avantage de leur disponibilité pour la répétition des mouvements par le patient, ainsi qu'une batterie de capteurs pour relever et enregistrer l'évolution des patients.Après une phase initiale de travail sur des robots manipulateurs, la recherche en thérapie assistée s'est rapidement orientée vers l'utilisation des exosquelettes. Ces dispositifs sont des systèmes robotisés se basant sur la structure du corps humain pour venir en complément de celle-ci. Les exosquelettes sont construits pour avoir leurs axes d'articulations correspondant à ceux d'un corps humain et pouvoir transmettre une assistance robotisée par différentes attaches. La présence de multiple effecteurs et la correspondance des axes permettent à l'exosquelette, contrairement au robot manipulateur, de cibler le membre souhaité pour coopérer et d'avoir un meilleur contrôle du couple appliqué à chaque articulation. Dans le travail thérapeutique, un meilleur contrôle de la posture du membre sera obtenu, et une plus grande variété d'exercices sera réalisable.Les problématiques abordées dans les travaux de cette thèse sont la commande d'exosquelette de membre supérieur droit pour la réalisation de mouvement dans le cadre de la rééducation et l'insertion de l'humain dans la boucle de commande par l'utilisation de signaux cérébraux. Cette thèse a été réalisée dans le laboratoire LISSI (Laboratoire Images, Signaux et Systèmes Intelligents) de l’UPEC.Le premier apport de ma thèse se situe dans la conception de lois de commande adaptative, basée sur de nouvelles lois de commutation permettant d'éviter à la commande toute singularité et garantissant la convergence de l'erreur vers zéro.Une seconde loi de commande a été proposée durant cette thèse, celle ci est basée sur l'estimation par réseaux neuronaux des paramètres dynamiques variants, qui sont dépendant de la morphologie de l'utilisateur. Pour garantir la stabilité de ce type de commande, une fusion avec une commande par mode glissant a été réalisée. En effet, cette commande connue pour sa robustesse permet de garantir la stabilité du système en compensant les éventuelles erreurs d'approximations des réseaux neuronaux.D'autre part, cette thèse a aussi permis la mise en place de travaux sur la fusion de ces lois de commande avec un traitement de signal EEG (électroencéphalographique). Ainsi à l'aide de ces signaux cérébraux, captés par un BCI ( brain computer interface), un algorithme a été établi pour permettre à l'utilisateur d'établir un choix de déplacement de l'exosquelette.

D’Tech Thesis SummaryThe importance of medical imaging as a core component of several medical application and healthcare diagnosis cannot be over emphasised. Integration of useful data acquired from different images is vital for proper analysis of information contained in the images under observation. For the integration process to be successful, a procedure referred to as image registration is necessary.The purpose of image registration is to align two images in order to find a geometric transformation that brings one image into the best possible spatial correspondence with another image by optimising a registration criterion. The two images are known as the target image and the source image. Image registration methods consist of having the two images referenced with control points. This is followed by a registration transformation that relates the two images and a similarity metric function that aims to measure the qualitative value of closeness or degree of fitness between the target image and the source image. Finally, an optimiser which seeks an optimal transformation inside the defined solution search space is performed.This research presents an automated image registration algorithm for solving multimodal image registration on lung Computer Tomography (CT) scan pairs, where a comparison between regular step gradient descent optimisation technique and evolutionary optimisation was investigated. The aim of this research is to carry out optimisation and performance evaluation of image registration techniques in order to provide medical specialists with estimation on how accurate and robust the registration process is. Lung CT scan pairs are registered using mutual information as a similarity measure, affine transformation and linear interpolation. In order to minimise the cost function, an optimiser, which seeks the optimal transformation inside the defined search space is applied.Determination of a transformation model that depends on transformation parameters and identification of similarity metric based on voxel intensity were carried out. By fitting transformation to control points, three transformation models were compared. Affine transformation produced the best recovered image when compared to non-reflective similarity and projective transformations. The results of this research compares well with documented results from EMPIRE 10 Challenge research and conforms to both theoretical principles as well as practical applications.The contribution of this research is its potential to increase the scientific understanding of image registration of anatomical body organs. It lays a basis for further research in performance evaluation of registration techniques and validation of procedures to other types of algorithms and image registration application areas, such as remote sensing, satellite communication, biomedical engineering, robotics, geographical information systems and mapping, among others, Résumé : Thèse de DoctoratL’importance de l’imagerie médicale comme élément principal dans plusieurs application médicales et diagnostiques de soin de santé est indispensable. L’intégration des données utiles obtenues des diverses images est vitale pour une analyse appropriée de l’information contenues dans ces images sous observation. Pour que ce processus d’intégration réussisse, une procédure appelée recalage d’image est nécessaire.Le but du recalage d’image consiste à aligner deux images afin de trouver une transformation géométrique qui place une des deux images dans la meilleure correspondance spatiale possible avec l’autre image en optimisant un critère de recalage. Les deux images sont dites image cible et image source. Les méthodes de recalage d’image consistent à avoir référencées par des points de contrôle. Ceci est suivi d’une transformation de recalage qui associe les deux images et d’une fonction définie sur base de la mesure de similarité qui a pour but de mesurer la valeur qualitative de proximité ou encore de degré de concordance entre l’image cible et l’image source. Enfin, un optimisateur qui cherche à trouver la transformation optimale au sein du champ de solution de la recherche, est appliqué.Cette recherche présente un algorithme automatique de recalage d’image dont le but est de résoudre le problème du recalage d’image à multiple modes sur une paire des clichés de tomographie par ordinateur (CT) faite sur les poumons. Dans cette méthode, une étude comparative entre la méthode classique d’optimisation par algorithme du gradient à pas fixe et celle de l’algorithme évolutionniste est menée. L’objectif de cette recherche est d’effectuer l’optimisation par des techniques de recalage d’image ainsi qu’évaluer la performance de ces mêmes techniques afin de doter les spécialistes du domaine médical d’une estimation de combien précis et robuste le processus de recalage serait. Les paires des clichés obtenues de la tomographie par ordinateur faite sur les poumons sont recalées en utilisant l’information mutuelle comme mesure de similarité, la transformation affine ainsi que l’interpolation linéaire. Un optimisateur qui recherche la transformation optimale au sein de l’espace de recherche est appliqué afin de minimiser la fonction économique (fonction objectif).Les études de détermination d’un modèle de transformation qui dépend des paramètres de transformation et de l’identification des mesures de similarité basée sur l’intensité du voxel ont été menées. En alignant la transformation avec les points de control, trois modèles de transformation sont comparés. La transformation affine produit la meilleure reconstitution d’image en comparaison aux transformations non réfléchissantes et projectives. Les résultats de cette recherche sont assez comparables à celles rapportées dans le challenge de recherche EMPIRE 10 et sont conformes à la fois aux principes théoriques aussi bien qu’aux applications pratiques.La contribution de cette recherche réside dans son potentiel à améliorer la compréhension scientifique du recalage d’image des organes anatomiques du corps humain. Cette recherche établie ainsi une base pour une recherche avancée sur l’évaluation de la performance des techniques de recalage et la validation des procédures sur d’autres types d’algorithmes et domaines d’application du recalage d’images comme la détection, la communication par satellite, l’ingénierie biomédicale, la robotique, les systèmes d'information géographique (SIG) et de localisation parmi tant d’autres

Navigation with Global Navigation Satellite Systems (GNSS) is a real challenge in harsh environments (suburban, urban, heavy foliage) due to multipath and signal blockage. This thesis proposes a number of GNSS receiver architectural and algorithmic solutions to deal with this challenge. These solutions aim at exploiting the strengths of scalar and vector tracking while minimizing their weaknesses and at utilizing the efficiency of some nonlinear Bayesian filtering techniques in addressing the nonlinearities and non-Gaussianities associated with the navigation and vector tracking problem. Attention is given to some Bayesian estimators that approximate the posterior distribution without linearizing the filtering model, namely the unscented Kalman and particle filtering methods, as well as to the extended Kalman filter, whose posterior estimation is grounded on linearization of the filtering model.First, a brief literature review that presents the fundamentals of GNSS and GNSS receivers together with the applied navigation and tracking algorithms is provided. Then an investigation of the GNSS receiver operation in multipath environments is performed. The thesis proposes models for characterizing multipath induced tracking errors in a vector tracking loop. These models make it possible to express the tracking errors with respect to multipath delay, multipath phase and multipath fading frequency. By exploiting the fact that multipath presence is mirrored on the Early-minus-Late correlator output, novel multipath detectors are devised. A correlator-based non-line-of-sight detector is designed as well. Attention is then directed towards the design of robust tracking and positioning GNSS receiver architectures that incorporate the proposed detectors among other signal quality indicators. A vector tracking scheme capable of detecting and excluding unhealthy measurements from position-velocity-time calculation in the navigator using correlator-based signal quality indicators is suggested. Two other novel tracking schemes, the adaptive scalar-vector tracking loop and the conjoint scalar-vector tracking loop, with the same fault detection and exclusion capability, are formulated. They benefit from vector tracking robustness in harsh environments and scalar tracking positioning accuracy in open sky environments. Experimental results show that the proposed solutions have better tracking and positioning performance than the usual scalar and vector tracking loops. Finally, the thesis presents a number of nonlinear Bayesian filtering approaches to solve the navigation and vector tracking problem. Iterative and adaptive strategies as applied to the unscented Kalman filter are studied. A novel unscented particle filter approach, the iterated adaptive unscented particle filter (IAUPF), is proposed. This approach exploits the convergence properties of iterative methods, the divergence suppression benefits of adaptive filters and the synergy of unscented Kalman and particle filtering approaches. Monte-Carlo simulations conducted with a posterior Cramér-Rao lower bound used as benchmarking reference as well as experimental results demonstrate that the IAUPF outperforms the other Bayesian estimators that are presented, La navigation avec les systèmes de navigation par satellites (GNSS) est un réel défi dans des environnements contraints (semi-urbain, urbain, feuillage dense) à cause des multitrajets et du masquage du signal. Cette thèse propose un nombre de solutions architecturales et algorithmiques pour le récepteur GNSS afin de pallier ces problèmes. Ces solutions se veulent capables d’exploiter les atouts des poursuites scalaire et vectorielle tout en minimisant leurs défauts et de profiter de l’efficacité de certaines techniques de filtrage Bayésien non linéaire quant à aborder la non-Gaussianité et les non-linéarités associées au problème de navigation et de poursuite vectorielle. Une attention particulière est accordée à certains estimateurs Bayésiens qui essaient d’approximer la loi a posteriori sans linéariser le modèle de filtrage, notamment le filtre de Kalman et les méthodes de filtrage particulaire, mais aussi au filtre de Kalman étendu, dont l’estimation de la loi a posteriori est basée sur la linéarisation du modèle de filtrage. En premier lieu, une brève étude bibliographique présentant les fondamentaux des systèmes et des récepteurs GNSS ainsi que les algorithmes de navigation et de poursuite y associés est faite. Ensuite le fonctionnement d’un récepteur GNSS en milieux contraints est investigué. La thèse propose des modèles pour caractériser les erreurs de poursuite induites par les multitrajets dans une boucle de poursuite vectorielle. Ces modèles permettent d’exprimer les erreurs de poursuite en fonction du délai, de la phase et de la fréquence d’évanouissement des multitrajets. En exploitant le fait que la présence des multitrajets se reflète sur la sortie moins des corrélateurs, de nouveaux détecteurs de multitrajets sont formulés. Un détecteur de masquage du signal direct est aussi proposé. L’attention se tourne ensuite vers la conception d’architectures robustes de poursuite et positionnement pour un récepteur GNSS, incorporant les détecteurs proposés et d’autres indicateurs de qualité. Une boucle de poursuite vectorielle capable de détecter et d’exclure des mesures qui ne sont pas saines du calcul de la solution de navigation en utilisant les indicateurs de qualité est proposée. Deux autres boucles de poursuite, la boucle de poursuite adaptative scalaire-vectorielle et la boucle de poursuite conjointe scalaire-vectorielle, avec la même capacité de détection et exclusion de fautes, sont formulées. Elles bénéficient de la robustesse de la poursuite vectorielle en milieux contraints et de la précision de la poursuite scalaire en milieux dégagés. Des résultats expérimentaux montrent que les solutions conçues offrent une meilleure alternative de poursuite et positionnement par rapport aux boucles usuelles de poursuite scalaire et de poursuite vectorielle. Enfin, la thèse présente des approches de filtrage Bayésien non linéaire pour résoudre le problème de navigation et de poursuite vectorielle. Des stratégies de filtrage itératives et adaptatives appliquées au filtre de Kalman sont étudiées. Une nouvelle approche de filtrage particulaire dénommée filtre particulaire itératif et adaptatif (IAUPF) est formulée. Cette approche exploite les propriétés de convergence des méthodes itératives, l’immunité à la divergence dont jouissent les filtres adaptatifs, et la synergie entre les approches de filtrage particulaire et de Kalman. Des simulations de Monte Carlo avec une borne inférieure de Cramér-Rao a posteriori comme référence ainsi que des résultats expérimentaux montrent que l’approche IAUPF a une meilleure performance comparativement aux autres estimateurs Bayésiens présentés

Although the progressive research and development of autonomous systems is fairly evident, such systems still require human interventions to solve the unforeseen complexities, and clear the uncertainties encountered in the execution of user-tasks. Thus, in spite of the system's autonomy, it may not be possible to absolutely disregard the operator's role. Human intervention, particularly in the control of auto-mobiles, may as well be hard to ignore because of the constantly changing operational context and the evolving nature of the drivers' needs and preferences. In order to execute the autonomous operations in conformity with the operator's expectations, it may be necessary to incorporate the advancing needs and behaviour of the operator in the design. This thesis formulates an operator behaviour model, and integrates the model in the control loop to adapt the functionality of a human-machine system to the operator's behaviour. The study focuses on a powered wheelchair, and contributes to the advancement of steering performance, through background assistance by modelling, empirical estimation and incorporation of the driver's steering behaviour into the control system. The formulation of the steering behaviour model is based on two fundamentals: the general empirical knowledge of wheelchair steering, and the experimental steering data captured by a standard powered wheelchair, on both virtual and real environments. The study considers a reactive directed potential field (DPF) method in the modelling of drivers' risk detection and avoidance behaviour, and applies the ordinary least square procedure in the identification of best-fitting driver parameters. The study also contributes to the development of a dynamic model of the wheelchair, usable under normal and non-normal conditions, by taking into consideration the conventional differential drive wheelchair structure with two front castor wheels. Derivation of the dynamic model, based on the Euler Lagrange formalism, is carried out in two folds: initially by considering the gravitational forces subjected to the wheelchair on inclined configurations with no slipping situations, and finally by incorporating slipping parameters into the model. Determination of the slipping parameters is approached from the geometric perspective, by considering the non-holonomic motions of the wheelchair in the Euclidean space. In the closed-loop model, the input-output feedback controller is proposed for the tracking of user inputs by torque compensation. The optimality of the resulting minimum-phase closed-loop system is then ensured through the performance index of the non-linear continuous-time generalised predictive control (GPC). Simulation results demonstrate the expected behaviour of the wheelchair dynamic model, the steering behaviour model and the assistive capability of the closed-loop system, Malgré le progrès en recherche et développement dans le domaine de système autonome, de tels systèmes nécessitent l’intervention humaine pour résoudre les problèmes imprévus durant l’exécution des tâches par l’utilisateur.Il est donc nécessaire, malgré cette autonomie, de tenir compte du comportement du conducteur et il est difficile d’ignorer l’effet de l’intervention humaine dans le cadre de l’évolution continue de l’environnement et des préférences de l’utilisateur. Afin d’exécuter les opérations selon les attentes de l’opérateur, il est nécessaire d’incorporer dans la commande les besoins de l’utilisateur.Dans les travaux présentés dans cette thèse un modèle comportemental de l’utilisateur est développé et intégré dans la boucle de commande afin d’adapter la commande à l’utilisateur. Ceci est appliqué à la commande des fauteuils électrique et assiste dans la navigation du fauteuil dans un milieu encombré.Le développement du modèle comportemental est basé sur la méthode de potentielles orientées et la détection des obstacles et le comportement du conducteur vs de ces obstacles par l’adaptation duL’étude contribue également au développement d’un modèle dynamique du fauteuil utilisable dans des situations normales et exceptionnelles telle que le dérapage. Ce modèle est développé pour un le cas le plus courant des fauteuil avec roues arrière conductrices utilisant le formalisme Euler Lagrange avec les forces gravitationnelles et sur des surfaces inclinées.Dans la formulation de la commande, le modèle du conducteur est introduit dans la boucle de commande. L’optimalité de la performance est assurée par l’utilisation du commande prédictif généralisé pour le système en temps continue. Les résultats de la simulation démontrent l’efficacité de l’approche proposée pour l’adaptation de la commande au comportement du conducteur

Wearable robotic system has become a well sought after mechanism in the field of biomechatronics engineering due to the the various possibilities it possess. These possibilities encompass the assistive and rehabilitative protocols rendered to disabled and elderly people, in order to enable them regain control of their limbs and of course increase the abilities of able-bodied persons. It therefore clearly drives the motive of bringing back paraplegics back on their feet as well as executing difficult task beyond human ability.Achieving the intended function of wearable robots requires the model dynamics of the physical system in relation to the tasks required to be performed by subjects. This demands a proper control measure which takes into account the safety of the wearer. For this purpose, bio-inspired control techniques and bounded nonlinear feedback controllers are considered. The latter control design ensures that the stipulated power required is not exceeded as well as the saturation of the actuator, while the former motivates the design of controllers based on the concept of Central Pattern Generators (CPG). CPGs are characterised as biological neural networks which can be represented by a set of coupled nonlinear oscillator situated in the spinal cord of mammals, having the capability of generating coordinated multidimensional rhythmic signals for the purpose of locomotion, under the control of simple input signals. These rhythmic signals are termed to be periodic or quasi-periodic in nature, hence performing this task in robotics and animal motor control has been a perpetual research problem. The movement of the lower limb of humans thus present a platform to investigate and address this difficulty.In this thesis, the analysis, simulation, and control of joints which relate to the human lower limbs via CPGs and feedback control techniques are investigated with an aim of practically implementing the control strategies using a lower limb exoskeleton is presented. To accomplish this goal, it is expedient to have comprehensive knowledge of the anatomy, physiology and the normal gait biomechanics of the human lower limbs. Understanding the theories, principles and mathematical background of nonlinear oscillators are also required. Control strategies using the inverse and the forward dynamics approach based on different types of coupled nonlinear oscillators and nonlinear feedback control techniques were considered for single/multiple degrees of freedom (DoF). Simulations and results were presented to verify the controller-human system ability to constantly and dynamically track and readapt its control parameters to maintain its desired motion dynamics, with reduced control torque values.This work basically deals with two distinct method of control systems; one which integrates bio-inspired methods with classical and nonlinear control techniques to govern the exoskeletons' joints with a human in the loop, and another which utilises bounded nonlinear feedback control techniques for same purpose, Les exosquelettes représentent des systèmes mécaniques portables qui ont reçu une grande attention de la part de la communauté scientifique ces derrières années, vues les possibilités qu'ils offrent.Ces possibilités concernent principalement les fonctions d'assistance et de réhabilitation des personnes en situation de handicape et personnes âgées, dans un objectif de leur permettre de recouvrir leur facultés motrices. Cependant, d'autres possibilités sont concernées comme permettre à des personnes paraplégiques de remarcher ou de permettre des opérations de manipulation excédent les capacités humaines.Pour permettre la réalisation des fonctions offertes par les robots portables, une connaissance fine de la dynamique du système est requise en relation avec les tâches à réaliser par les sujets. Par ailleurs, des approches de commande sûres qui prennent en compte la sécurité des usagers est nécessaire. Dans cet objectif, des techniques de commande bio-inspirées avec des techniques de commande par découplage non-linéaire sont considérées. Les dernières assurent que la loi de commande est stable et bornée en prenant en considération le domaine de saturation des actionneurs alors que les premières ont inspiré la conception de contrôleurs basés sur les oscillateurs locaux non-linéaires (Central Pattern Generators : CPG).Les CPGs sont modélisés par des réseaux de neurones qui peuvent être représentés par un ensemble d'oscillateurs non-linéaires situés dans la moelle épinière, avec des capacités de génération de signaux rythmiques multidimensionnels synchrones pour remplir des fonctions motrices sous le contrôle de simples signaux de commande. Ces signaux sont supposés être de nature périodique ou semi-périodique, dont la génération pour étudier les systèmes de locmotion humain reste un problème de recherche d'actualité.Dans la présente thèse, l'analyse, la simulation et la commandes des articulations d'un robot portable utilisé pour les membres inférieurs en utilisant, d'un côté les oscillateurs locaux non-linéaires et d'un autre côté des techniques de commande par découplage non-linéaire sont proposés, avec comme objectif final de permtre la mise en œuvre des approches proposées sur la plate-forme expérimentale développée au sein du FSATI (French South African Institute fo Technology).Pour atteindre l'objectif qui a été fixé par les travaux de recherche engagés, une étude de l'état de l'art sur les aspects liés à la connaissance de l'anatomie, la physiologie et l'analyse biomécanique de la marche humaine a été effectuée. Par ailleurs, une étude détaillée des oscillateurs locaux non-linéaires en parallèle avec les approches de commande directe et inverse, ont permis la proposition de stratégies de commande qui couplent les oscillateurs non-linéaires d'un côté et des techniques de découplage non-linéaire d'un autre côté ont été proposées et validées sur des systèmes de plusieurs degrés de liberté. Des simulations intensives ont été conduites afin de vérifier la capacité d'adaptation temps des approches de commandes mises en œuvre avec l'humain ans la boucle.Les contributions de la présente étude concerne deux approches de commande. La première approche concerne l'intégration d'une approche bio-inspirée, basée sur les oscillateurs locaux non-linéaires et la deuxième approche est basée sur les techniques de commande bornée par découplage non-linéaire

Taking into account high oil prices and environmental awareness, the development of electric vehicles (EVs) is considered as a healthier mode of transportation. Amongst other eco-friendly vehicles, EVs are considered as a favourable solution for a greener energy because the electricity they consume can be generated from a wide range of sources which include fossil fuel, nuclear power and renewable energy. However, users and owners of EVs feel uncomfortable because EVs require sufficient electrical energy battery storage on-board to provide sufficient driving autonomy.Lumped inductive power transfer (LIPT) system is a new technology that allows the transfer of electric power between its air-cored primary and secondary coils via high frequency magnetic fields to a consuming device. Unlike the conventional plug-in system, LIPT system is capable of providing a safe, efficient and convenient overnight recharging of EVs. However, its main limiting factor is the poor performance of its magnetic coupling structure (MCS), which is intended to transfer power efficiently. Thus the problem statement of this thesis is to improve the performance of MCS models for LIPT systems.Firstly, based on a more efficient and relevant mathematical model available in the literature, MATLAB code is implemented to compute the mutual inductance between air-cored filamentary circular (FC) coils. Also, the computation and experimental validation of the magnetic fields between two FC coils are presented.Furthermore, computational models of an IPT system for EV battery charge are presented in this thesis. Based on the technical specifications of Renault ZOE, the results obtained show that by supplying a higher frequency AC voltage to the primary coil of the MCS and compensating the primary and secondary sides of the air-cored coils with series-series capacitors, the 3 kW single-phase and 22 kW three-phase IPT systems modelled using MATLAB/Simulink are capable of delivering the electricity needed to power the Renault ZOE.Finally, in order to recommend a suitable and cost-efficient MCS model that can help transfer electric power more efficiently for the battery charging of EVs and E-bikes, a 3-D finite element analysis (FEA) package called COMSOL multiphysics is used to design, compute and investigate a more complex and realistic MCS model of LIPT systems. The designed MCS models incorporate air-cored coils with proper configuration of magnetic cores (e.g. ferrite), structural steel covering for the bottom part of the primary coil and top part of the secondary coil and lastly, iron plate which serves as a covering for the primary coil installed underground and the chassis or underbody structure of EVs. The performance of the designed models are determined by the values of the mutual inductance and induced voltage obtained from COMSOL., Compte tenu du contexte économique du marché des hydrocarbures et les problématiques environnementales, le développement des véhicules électriques (VE) prend de l’ampleur car ils sont considérés comme plus écologiques. Aujourd’hui, les véhicules électriques sont considérés comme une solution favorable pour une énergie plus verte. L'électricité qu'ils consomment peut être générée à partir d'un large éventail de sources qui comprennent les combustibles fossiles, l'énergie nucléaire et les énergies renouvelables. Toutefois, les utilisateurs et les propriétaires de véhicules électriques ont encore des réticences car cela nécessite un stockage d'énergie électrique à bord pour assurer une bonne autonomie.Le système de transfert de puissance par effet inductif (LIPT en anglais) est une nouvelle technologie qui permet le transfert d'énergie électrique par champ magnétique et un système de bobines primaires et secondaires. Le champ magnétique est un champ haute-fréquence à plusieurs dizaines de kilohertz. Par rapport au système de câble conventionnel, le système LIPT est capable de fournir une recharge qui est pratique mais également efficace des véhicules électriques. Cependant, actuellement son principal facteur limitant est la mauvaise performance de sa structure de couplage magnétique (MCS). L’objectif de cette thèse est d'améliorer la performance des systèmes MCS pour les systèmes de LIPT afin de concevoir des systèmes à meilleur rendement.Dans un premier temps, sur la base de modèles mathématiques issus de la littérature, un code Matlab a été mis en œuvre pour calculer l'inductance mutuelle des systèmes de bobines mise en jeu dans le MCS. Puis, le calcul et la validation expérimentale des champs magnétiques entre le primaire et le secondaire a été effectué.Dans un second temps, un modèle d'un système LIPT pour la charge d’une batterie de véhicule électrique est présenté. Sur la base des spécifications techniques d’une Renault ZOE, les résultats obtenus montrent que, en adaptant la fréquence de la bobine primaire et en compensant avec un système série-série de condensateurs, un système à 3 kW et un système à 22 kW peuvent atteindre des performances permettant la recharge d’une Renault Zoe dans de bonnes conditions.Enfin, une analyse par éléments finis (FEA) sous COMSOL est développée pour la conception, le calcul et l’optimisation de systèmes MCS plus complexes de nouveaux LIPT. Les modèles de MCS conçus intègrent des bobines d'air évidées avec des configurations appropriées de noyaux magnétiques (par exemple en ferrite), avec des études également sur des parties couvrantes des bobines primaires et secondaires en acier. Les performances des modèles conçus sont déterminées par les valeurs de l'inductance mutuelle et la tension induite qui sont deux critères d’évaluations.

Vehicular ad-hoc Networks (VANETs) are becoming more and more popular as a means to increase the traffic safety and comfort. VANETs have some characteristics due to the vehicle's high speeds mobility such as rapid changes of topology, potentially large-scale, veritable network density and so on. These characteristics have important implications for design of MAC protocol in VANETs. 802.11p standard are attracting increasingly more attention as an important part of the WAVE protocol in VANETs. However the QoS of 802.11p standard can not yet meet the requirements of the realistic vehicular communication networks. Three sub-problems are identified in this study: (i) CCH and SCHs duty cycle problem, (ii) QoS analysis problem and (iii) Multichannel MAC protocol designing problem. Based on these problems, three approaches are proposed in this study: (i) To propose a duty cycle adaptive MAC protocol in which the duty cycle of CCH and SCH will be adapted based on the realtime network traffic; (ii) To propose a model to analyze the QoS in VANETs with various SCH interval settings; (iii) To propose a scheduling and QoS enhancement multi-channel MAC protocol in VANETs. All the proposed algorithms are implemented and validated in discrete event simulator. The simulation results demonstrate that the important QoS metrics such as the reliability, throughput, successful throughput, network capacity and channel utilization are improved by the proposed algorithms in this project, Les réseaux véhiculaires (VANETS) sont en phase de devenir incontournable pour les infrastructures routière améliorant le confort et la sécurité. Les VANETS possèdent quelques caractéristiques, liées la vitesse des véhicules qui peut devenir importante, concernant le changement rapide de topologie la mise à l'échelle, la densité du réseau, etc. Ces caractéristiques ont d'importantes implications sur la conception des protocoles d'accès (MAC) dans ce type de réseau. Le standard 802.11p est l'objet d'une attention importante par la communauté comme partie important du standard WAVE pour les VANETs. Cependant, la gestion de la qualité de service par le standard 802.11p ne peut pas répondre aux exigences d'une communication réaliste dans les réseaux véhiculaires. Trois sous problèmes ont été identifiés pour être traités dans le cadre des travaux de la présente thèse:(i) Problème du rapport cyclique des canaux CCH et SCH(ii) Problème de la qualité de service(iii) Problème de conception du protocole MAC multi-canaux. En se basant sur ces trois problèmes, trois approches ont été proposées(i) Une approche adaptative des rapports cycliques a été proposée, basée sur le trafic-réseau temps-réel(ii) Un modèle de la qualité de service a été proposé pour analyser la QdS dans les VANETS sous différentes configuration de l'intervalle SCH(iii) Un mécanisme d'ordonnancement ainsi qu'un protocole MAC multi-canaux a été proposé. Les différentes propositions ont été validées en simulation en exploitant différentes métriques de QdS pour montrer l'intérêt des solutions apportés aux problèmes identifiés ci-dessus

The thesis addresses the improvements of the Standard Hough Transform (SHT) for image processing applications. As a common tool for straight line segment detection, the Hough Transform (HT) has received extensive attention since its proposal in 1962. HT is robust since it converts the straight line detection problem to a peak seeking problem in the parameter space (also called HT space or Hough space). Feature points in the image space are mapped to the parameter space, according to the normal formulation of the possible straight lines crossing them. The cells in the digitalised parameter space obtain votes from the feature points. The local maxima, corresponding to peaks are built when corresponding cells obtain more votes than the ones around them. The peaks detected in the parameter space are then mapped back to the image space for validation. Unfortunately, when mapping feature points in the image space to the parameter space in conjunction with the voting process, rounding operations are employed, which leads to several problems of HT. The robustness, accuracy and resolution are all affected. This thesis aims to solve these problems, and the following contributions were made towards this goal : A) Because of the resolution limitation of SHT, the topics of how to select a “good” resolution, how to extend the resolutions of SHT and how to employ the super-resolution technique in HT are covered. In the research of these topics, several outputs are obtained, which are helpful to image processing applications. These include:- The map of HT performance versus resolutions is drawn, according to which “good” choices of resolutions can be found.- HT resolutions are extended by geometrical analysis of the HT butterflies.- Super resolution HT is proposed with consideration to the features of HT.- Self-similarity of the HT butterflies is discovered and employed to obtain high resolution HT with high reliability. B) For the accuracy defect of SHT, the error system of HT is studied when the image is shifted in the image space. The detection errors are employed to compensate for the defect, and an accurate HT is proposed. In order to evaluate existing HT varieties, an ideal HT is proposed as a standard, Les travaux effectués dans le cadre de la présente thèse concernent l'analyse et les améliorations apportées à la transformée de Hough Standard (SHT), utilisée en traitement d'image comme simple outil de détection de segments de lignes droites. La transformée de Hough a reçu, depuis sa proposition en 1962, une attention particulière de la part de la communauté. La HT est considérée comme une méthode robuste, dont le principe repose sur la transformation le problème initial de détection de segments de lignes droites en un problème de section de sommets dans l'espace des paramètres, appelé aussi espace HT ou espace de Hough. Les points candidats dans l'espace image sont mis en correspondance points dans l'espace de Hough, en utilisant Le principe avancé par la transformée de Hough est qu'il existe un nombre infini de lignes qui passent par un point, dont la seule différence est l'orientation (l'angle). La transformée de Hough permet de déterminer lesquelles de ces lignes passent au plus près du domaine d'intérêt. Les cellules dans l'espace de Hough échantillonné obtiennent des votes des points candidats. Les maxima locaux, correspondant aux sommets sont construit lorsque les cellules considérées obtiennent plus de votes que les cellules voisines. Les sommets détectés alors dans l'espace des paramètres sont transformée dans l'espace image pour validation. Malheureusement, les opérations les opérations de transformation directe, de l'espace image vers l'espace des paramètres, et inverse engendrent des opérations d'approximation, ce qui est la source de plusieurs problèmes de la transformée de Hough, qui affectent les aspects de robustesse, précision et résolution. On se propose de résoudre ces problèmes dans le cadre des travaux engagés dans le cadre de la thèse. Les contributions, détaillées ci-dessous, ont pu être proposées. A) Pour adresser le problème de limitation en termes de résolution de la SHT, les points concernent la sélection d'une bonne résolution, l'extension de la résolution de la SHT et l'utilisation des techniques de super-résolution pour la HT ont été couverts et de nouvelles propositions ont été faites et qui sont d'une utilité certaine pour les applications de traitement d'image.- la relation entre la performance de la HT et la résolution est proposée, ce qui permet de garantir le bon choix.- Une technique de super-résolution ets proposée en s'appuyant sur le principe de la HT- Une auto-similarité dans les échantillons HT a été découverte et a été utilisée pour obtenir une résolution supérieure de la HT avec un grande fidélité. B) Pour adresser le problème de la précision de la SHT, les erreurs de la HT ont été analysées, lorsque l'on fait subir des transformations géométriques à l'image source. Les erreurs ainsi détectées ont été utilisées pour compenser le manque de précision de la SHT, aboutissant ainsi à une HT plus précise. Afin de permettre l'évaluation de performance des approches proposées dan sel cadre de la thèse, une transformée de Hough idéale est proposée comme référence

Ambient Intelligence is considered as the major application domain of ubiquitous computing. It aims at implementing intelligent environments to improve the daily activities, the well-being and the safety of users. An ambient environment has a multitude of devices (sensors, actuators), providing atomic ubiquitous services which are distributed and heterogeneous. These atomic services available in these environments, do not directly or fully meet all the requirements and the needs of users. Moreover, these high-level services may involve the interaction of ubiquitous services owned and/or controlled by different structures or domains (i.e. multi-domain environment). In this work, we studied the problems related to the aggregation of atomic ubiquitous services in cooperations and compositions in an ambient multi-domain environment in order to meet users' needs. This study concerns the semantic and behavioral interoperability of high-level composite services and their formal specification. This latter, is based on the Basic Constructive Description Logic, namely, BCDL0 formal model. This work contributes by realizing a semantic framework for the design and formal verification of cooperative processes in the interactive theorem prover Isabelle/HOL. The proposed semantic framework consists of an extensible cooperation ontology, a sound formal specification language of ubiquitous services and a components for mapping the formally proved cooperatives processes to the targeted infrastructures, L'intelligence ambiante est considérée comme l'une des évolutions majeures de l'informatique ubiquitaire. Elle vise la mise en œuvre des environnements ambiants ou des espaces intelligents permettant d'améliorer la vie quotidienne des utilisateurs, leur bien-être et leur sécurité. Un environnement ambiant dispose d'une multitude d'équipements (capteurs, actionneurs), fournissant des services ubiquitaires atomiques qui sont distribués et de nature hétérogène. Ces services atomiques disponibles dans ce type d'environnement, ne répondent pas directement ou complètement à toutes les exigences et les besoins des utilisateurs. Par ailleurs, ces services de haut niveau peuvent impliquer l'interaction de services ubiquitaires appartenant et/ou contrôlés par différentes structures ou domaines (i.e. environnement multi-domaine). Dans ces travaux de thèse, on a étudié les problématiques reliées à l'agrégation des services ubiquitaires atomiques dans des coopérations et des compositions dans un environnement ambiant multi-domaine afin de répondre à des besoins utilisateurs. Cette étude concerne l'interopérabilité sémantique et comportementale des compositions de services ainsi que leurs spécifications dans un modèle formel basé sur la logique constructive BCDL0. Cette thèse a contribuée par la réalisation d'un canevas sémantique permettant la conception et la vérification formelles des processus coopératifs dans l'assistant à la preuve Isabelle/HOL. Le canevas sémantique proposé consiste en une ontologie de coopération extensible et un langage de spécification de services fiable (Sound) ainsi que des modules de mise en correspondance vers des infrastructures cibles

In this thesis three themes related to wireless sensor networks (WSNs) are covered. The first one concerns the power loss in a node signal due to voltage droop in the battery of the node. In the first part of the thesis a method is proposed to compensate for the apparent increase in the calculated distance between the related nodes due to decrease in the energy of the signal sending node battery. A function is proposed whose arguments are the apparently observed RSS and the current voltage of the emitter node battery. The return of the function is the corrected RSS that corresponds to the actual distance amongst the connected nodes. Hence increasing the efficiency of the RSS based localization methods in WSNs. In the second part of the thesis a position estimation method for localization of nodes in a WSN is proposed. In the proposed localization algorithm anchor nodes are used as landmark points. The localization method proposed here does not require any constraint on the placement of the anchors; rather any three randomly chosen nodes can serve as anchors. A heuristic approach is used to find the relative topology with the help of distance matrix. The purpose of the distance matrix is to indicate whether or not a pair of nodes has a connection between them and in case of connectivity it gives the estimated distance between the nodes. By using the information of connectivity between the nodes and their respective distances the topology of the nodes is calculated. This method is heuristic because it uses the point solution from the intersection of two circles instead of conventional triangulation method, where a system of three quadratic equations in two variables is used whereby the computational complexity of the position estimation method is increased. When two connected nodes have another node in common, then by using the information of distances between these interconnected nodes, two possible positions are calculated for the third node. The presence or absence of a connection between the third node and a fourth node helps in finding the accurate possibility out of the two. This process is iterated till all the nodes have been relatively placed. Once the relative topology has been calculated, we need to find the exact symmetry, orientation, and position of this topology in the plane. It is at this moment the knowledge of three nodes positions comes into action. From the relative topology we know the temporary coordinates of the nodes. By having a comparison of certain characteristics between the temporary coordinates and the exact coordinates; first the symmetry of relative topology is obtained that would correspond to the original topology. In other words it tells whether or not the relative topology is a mirror image of the original topology. Some geometrical operators are used to correct the topology position and orientation. Thus, all the nodes in the WSN are localized using exactly three anchors. The last part of the thesis focuses on the detection of faults in a WSN. There is always a possibility that a sensor of a node is not giving accurate measurements all of the time. Therefore, it is necessary to find if a node has developed a faulty sensor. With the precise information about the sensor health, one can determine the extent of reliance on its sensor measurement. To equip a node with multiple sensors is not an economical solution. Thus the sensor measurements of a node are modeled with the help of the fuzzy inference system (FIS). For each node, both recurrent and non-recurrent systems are used to model its sensor measurement. An FIS for a particular node is trained with input variables as the actual sensor measurements of the neighbor nodes and with output variable as the real sensor measurements of that node. The difference between the FIS approximated value and the actual measurement of the sensor is used as an indication for whether or not to declare a node as faulty, Dans cette thèse, on s'est intéressé à trois problématiques des réseaux de capteurs sans fil (WSN). Dans un premier temps nous avons analysé l'impact de la chute de tension dans la batterie du nœud sur la puissance du signal en réception. On propose alors une méthode pour compenser l'augmentation apparente de la distance calculée entre les nœuds due à la diminution de l'énergie de la batterie. Pour les nœuds passant par deux états principaux endormi et actif, on propose d'étudier, la relation entre la diminution de la tension de la batterie en fonction du temps passé par un nœud dans l'état actif. Ensuite, on calcule le rapport entre la RSS et la distance entre les nœuds connectés avec des batteries complètement chargées. Après on mesure la RSS en faisant varier la tension de la batterie du nœud émetteur et en gardant le nœud récepteur à une distance constante. Finalement, on propose une relation entre la RSS observée et la tension actuelle de la batterie du nœud émetteur. Cette fonction permet de calculer la valeur corrigée de la RSS qui correspond à la distance réelle entre les nœuds connectés. Ainsi l'efficacité des méthodes de la localisation basée sur la RSS se trouvent améliorées. Dans la deuxième partie de cette thèse on propose une méthode d'estimation des positions des nœuds dans un WSN. Dans l'algorithme de localisation proposé, on utilise des nœuds ancres comme des points de référence. On a utilisé une approche heuristique pour trouver la topologie relative avec l'aide de la matrice de distance. Le but de la matrice de distance est d'indiquer s'il existe une connexion entre une paire de nœuds donnée et en cas de connectivité, la distance estimée entre ces nœuds. En utilisant les informations de connectivité entre les nœuds et leurs distances, on obtient la topologie du réseau. La méthode proposée utilise la solution de l'intersection de deux cercles au lieu de la méthode classique de triangulation, où un système quadratique de trois équations avec deux variables est utilisé ce qui rend la complexité de calcul augmentée. Lorsque deux nœuds connectés ont un autre nœud en commun, puis en utilisant les informations de distances entre ces nœuds interconnectés, nous pouvons calculer deux positions possibles pour le troisième nœud. La présence ou l'absence d'un lien entre le troisième nœud et un quatrième nœud, permet de trouver la position précise. Ce processus est réitéré jusqu'à ce que toutes les positions des nœuds aient été obtenues. Une fois la topologie relative calculée, il faut trouver la symétrie, l'orientation et la position de cette topologie dans le plan. C'est à ce moment que la connaissance des positions des trois nœuds entre en action. La topologie donne les coordonnées temporaires des nœuds. En ayant une comparaison de certaines caractéristiques entre les coordonnées temporaires et les coordonnées exactes, on trouve d'abord la symétrie de la topologie relative qui correspondrait à la topologie originale. En d'autres termes on vérifie si oui ou non la topologie relative est une image miroir de la topologie originale. Des opérateurs géométriques sont alors utilisés pour corriger la topologie relative par rapport à la topologie réelle. Ainsi, on localise tous les nœuds dans un WSN en utilisant exactement trois ancres. Dans la dernière partie de cette thèse, on propose une méthode pour la détection de défauts dans un WSN. Il y a toujours une possibilité qu'un capteur d'un nœud ne donne pas toujours des mesures précises. On utilise des systèmes récurrents et non récurrents pour la modélisation et on prend comme variable d'entrée, en plus des variables du nœud en question, les informations des capteurs voisins. La différence entre la valeur estimée et celle mesurée est utilisée pour déterminer la possibilité de défaillance d'un nœud

This thesis proposes a new approach relating to the packets scheduling algorithm, the cooperation scheme and the nodes' mobility for IEEE 802.11 wireless network family. Considering the QoS delivery process at the MAC level consists the main feature of the proposal research study. The analysis of the current mechanisms leads to keep the protocol EDCF as the basic model for our work platform. In order to overcome the weakness of the earlier 802.11 standard, a new model based on multi-scheduler algorithm, called HCF-T, is proposed. The achieved performances are summarized following several criteria: traffic control, throughput improving, collisions avoidance, and network load decreasing. Furthermore, in the way of getting better results according to the PHY layer, we presented and analyzed a model of cooperative diversity scheme. It included a couple of relaying protocols AAF and DAF supported by a set of combining techniques to backup the signal at the receiver. To support node's mobility within EDCF, we built-up a model of WLAN which able to track node motion and control the access as in real condition. In the case of overlapping APs ranges, we developed a Multi-coverage algorithm aiming to carry out the session associations. As a result, three levels (Low, Medium, and High) of node's speed are discerned. Finally, EDCF has been implemented on various static and dynamic scenarios. The performances, based on the main MAC-layer metrics, such as throughput, End-2-End delay, and jitter, have been classified and comprehensively evaluated, Cette thèse a pour objectif de proposer de nouvelles approches d'ordonnancement, de coopération et de gestion de la mobilité dans les réseaux sans fil de type IEEE 802.11. Le maintien de la qualité de service (QoS), au niveau MAC, représente la caractéristique fondamentale de ces approches. L'analyse des mécanismes existants nous a conduits à retenir le protocole EDCF, supportant la QoS, comme une base de travail pour l'ensemble de nos propositions. Dans le but de pallier certaines faiblesses du standard 802.11, une nouvelle architecture à base de multi-ordonnanceurs HCF-T, est proposée. Les performances obtenues sont exprimées en termes de gestion du trafic, de maintien du débit, d'élimination de collisions et de réduction de la charge du réseau. Ensuite, un schéma coopératif est présenté et analysé. Il comporte une étude de deux protocoles de relayage AAF et DAF ainsi qu'une évaluation d'un ensemble de techniques de combinaison au niveau du récepteur. Concernant la problématique de la mobilité, nous avons retenu et analysé un scénario prenant en considération les différentes situations rencontrées dans un modèle réel. Un algorithme de résolution multi-couvertures est proposé afin de traiter l'accès dans les zones de recouvrement. Cette étude a mené à distinguer trois régimes de mobilité : faible, moyen et fort. Les performances sont évaluées en fonction des métriques MAC et pour chaque mode de mobilité, un schéma de synthèse est établi

In this thesis, we are interested in a micro-robot study for the implementation of a mini-invasivesurgery technique. The medical application concerns the treatment of the artery abdominal aorta aneurysms. This micro-robot located at the extremity of the catheter permits thecatheter movements into the artery avoiding minimizing contacts with the artery walls. The system can be used for the surgical gesture learning and for the surgeon assistance during the medical operation. Because of its structure and its physical properties the micro-robot is considered as a continuum robot. It is composed of one or several elementary modules. In our study we consider each elementary module as a parallel robot. Then the inverse kinematics model has been established by using techniques of parallel robotics. The proposed modelling approach allows the expression of the model according to the micro-robot geometric parameters. A study on the identification of these parameters has been developed by an inverse geometric modelcalibration. The given simulation results validate the developed models on the one hand, the proposed calibration method on the other hand. We have developed a three-dimensional simulator integrating the model of an artery segment, the micro-robot model, and a joystick with force feedback. The implementation of active, planned and guided navigation on this simulator allows to constrain the surgeon gestures during the movements of the micro-robot inside the artery, Dans cette thèse, nous nous sommes intéressés à l 'étude d'un micro-robot destiné à la mise en oeuvre d'une technique de chirurgie mini-invasive pour le traitement des anévrismes de l'artère aorte abdominale. Ce micro-robot placé à l'extrémité d'un cathéter, rendant ce dernier actif, permettra la navigation à l'intérieur de l'artère en évitant les contacts avec les parois de celle-ci. Le système sera destiné à l'apprentissage du geste chirurgical et à l'assistance du chirurgienpendant l'opération. De par sa structure et ses propriétés physiques, le micro-robot, pouvantêtre composé de plusieurs modules élémentaires, entre dans la catégorie des robots continuum. Dans notre étude, un module élémentaire est considéré comme étant un robot parallèle. Lesmodèles géométriques et cinématiques inverses ont alors été établis en utilisant les techniques dela robotique parallèle. L'approche de modélisation proposée permet de faire ressortir explicitement du modèle les paramètres géométriques du micro-robot. Une étude sur l'identificationde ces paramètres a été effectuée par le calibrage du modèle géométrique inverse. Des résultatsde simulation sont présentés validant d'une part les modèles développés et d'autre part la méthode de calibrage proposée. Afin de mettre nos modèles en situation, nous avons développé unsimulateur tridimensionnel intégrant le modèle d'un segment de l'artère, le modèle du micro-robotet un syntaxeur à retour de force. La mise en place d'une navigation active, planifiée etguidée dans ce simulateur permet de contraindre les gestes du chirurgien lors de la navigation du micro-robot à l'intérieur de l'artère

This dissertation endeavors to contribute enhancements in goodputsof the IEEE 802.11-based Wireless Multi-hop Networks (WMhNs).By performing exhaustive simulations, for the deep analysis and detailed assessment of both reactive (AODV, DSR, DYMO) and proactive (DSDV, FSR, OLSR) protocols for varying mobilities, speeds, network loads and scalabilities, it is observed that a routing link metric is a significant component of a routing protocol. In addition to finding all available paths, the fastest end-to-end route is selected by a link metric for the routing protocol. This study aims the quality routing. In the class of quality link metrics, Expected Transmission Count (ETX) is extensively used. Thus, the most recently proposed ETX-based metrics have been analyzed. Though, newly developed metrics over perform ETX but still they can be improved. By profound analysis and particularized comparison of routing protocols depending upon their classes (reactive and proactive) and ETX-based metrics, we come to realize that users always demand proficient networks. In fact, WMhNs are facing several troubles which they expect to be resolved by the routing protocol operating them. Consequently, the protocol depends upon the link metric for providing quality paths. So, we identify and analyze the requirements to design a new routing link metric for WMhNs. Because, considering these requirements, when a link metric is proposed, then : firstly, both the design and implementation of the link metric with a routing protocol become easy. Secondly, the underlying network issues can easily be tackled. Thirdly, an appreciable performance of the network is guaranteed. Keeping in view the issues of WMhNs, increasing demands of users and capabilities of routing protocols, we propose and implement a new quality link metric, Interference and Bandwidth Adjusted ETX (IBETX). As, MAC layer affects the link performance and consequently the route quality, the metric therefore, tackles the issue by achieving twofold MAC-awareness. Firstly, interference is calculated using cross-layered approach by sending probes to MAC layer. Secondly, the nominal bit rate information is provided to all nodes in the same contention domain by considering the bandwidth sharing mechanism of 802.11. Like ETX, our metric also calculates link delivery ratios that directly affect throughput and selects those routes that bypass dense regions in the network. Simulation results by NS-2 show that IBETX gives 19% higher through put than ETX and 10% higher than Expected Throughput (ETP). Our metric also succeeds to reduce average end-to-end delay up to 16% less than Expected Link Performance (ELP) and 24% less than ETX, Les travaux de recherche menés dans le cadre de cette thèse concernent l'amélioration du débit et de la qualité de service dans les réseaux sans fil basés sur les standards de la famille 802.11. Des simulations exhaustives ont été menées pour l'analyse et l'évaluation des performances des protocoles de routages réactifs AODV, DSR et DYMO ainsi que des protocoles de routage proactifs DSDV, FSR et OLSR. Plusieurs paramètres ont été considérés comme la mobilité des noeuds la charge du réseau et la mise à l'échelle. Nous pouvons observer que la métrique de qualité de lien est un paramètre important dans toute stratégie de routage. L'objectif du présent travail est de proposer une analyse comparative des différents protocoles de routage, basée sur les métriques de qualité de lien et de proposer une nouvelle métrique permettant d'améliorer le routage en termes de routage et de délai bout-en-bout. Dans la disparité des métriques proposées dans la littéraire, la métrique ETX (Expected Transmission Count) a été largement utilisée. Aussi, nous focaliserons, principalement, sur les métriques basées sur ETX. Notre constatation est que les besoins en termes de qualité de service dans les réseaux sans fil multi-saut, sont affecté entre autre par le protocole de routage adopté et la métrique de lien utilisée. Cependant, ceci est aussi strictement lié à la qualité des canaux de communication au niveau physique et l'état des files d'attente au niveau MAC. Aussi, une nouvelle métrique de qualité de lien est proposée, basée sur ETX et appelée IBETX (Interference and Bandwidth Adjusted ETX). Cette nouvelle métrique prend en considération les effets de l'interférence et de l'impact de l'état de la couche MAC sur les performances des liens de communication. Des simulations ont été conduites sous NS-2 afin de montrer l'intérêt de la métrique utilisée. Ainsi IBETX améliore le débit efficace de 19% par rapport à ETX et de 10% par rapport à ETP (Expected Throughput). Par ailleurs, la métrique proposée réduit le délai bout en bout de 16% par rapport à ELP ( Expected Link Performance) et de 24% par à ETX