Le vieillissement de la population sans doute catalysera l’augmentation des maladies chroniques et intensifiera le besoin de solutions d’assistance à la personne. Pendant que les chercheurs s’activent à apporter des réponses aux problèmes de santé publique qui s’accentuent, en s’appuyant sur les technologies de l’information et de la communication, le nombre des objets connectés connait une expansion fulgurante. Ainsi, le désir de révolution des technologies pour la santé, afin de faire face à la menace pathologique, coïncide avec le développement de l’Internet des objets 1. En effet, grâce aux innovations technologiques et au progrès médical, nombre de pathologies, souvent chroniques pourraient être suivies en temps réel et en tout lieu. Dans ce contexte, la gestion ou le partage des ressources de communication, la compatibilité des technologies et les performances à atteindre constituent des défis importants. Cet accroissement significatif du volume des communications, les contraintes de mobilité imposées par le contexte du suivi de patient ainsi que les besoins de qualité dans les transmissions de données médicales, révèlent une aspiration à des infrastructures de communication plus flexibles.Dans cette thèse, nous présentons une architecture de communication basée sur les réseaux Radio Cognitive pour répondre à cette exigence. Le caractère adaptable, flexible et autonome de la solution proposée permet d’aspirer à de meilleures performances. Ainsi, pour l’évaluation de son efficacité,nous avons choisi d’analyser et de tester trois critères importants pour les transmissions de données médicales urgentes.La connectivité en tout lieu : Ce premier critère est essentiel dans la mesure des performances et l’estimation de la fiabilité d’une infrastructure réseau dédiée à la santé. Plus précisément, toute solution de communication envisagée, doit être en mesure d’accompagner le patient suivi dans son environnement. En effet, la haute disponibilité des services réseaux et la qualité offerte sont déterminantes pour le suivi de patient à distance. Nous proposons dans cette première contribution, un mécanisme de prédiction spectrale capable d’examiner l’état d’occupation des bandes de fréquence. Cet algorithme associé au module de prise de décision Radio Cognitive, permet de parer aux éventuelles discontinuités de connexion réseaux.La gestion des interférences : Il s’agit du second critère qui évalue le degré de coexistence des ondes garantit par l’architecture, dans un contexte de prolifération des réseaux et des objets connectés. Le matériel communicant doit être capable de percevoir, d’analyser son environnement et d’agir en fonction des différentes contraintes. L’intérêt étant de protéger le matériel surtout médical, souvent très sensible aux bruits. Le suivi du patient devient alors possible à domicile ou à l’hôpital par exemple, avec un niveau d’interférence acceptable. Ainsi, tout en proposant un modèle de déploiement du réseau Radio Cognitive dans un centre hospitalier, nous définissons des exemples de fonctions permettant une adaptation dynamique des paramètres de communication en fonction de la sensibilité des équipements médicaux de proximité.L’efficacité dans la transmission de contenu multimédia : Ce dernier critère symbolise la capacité de l’architecture à fournir du contenu de qualité pour une assistance en temps réel. En effet, un réseau de soin à domicile ou une situation d’urgence peut nécessiter la transmission d’images ou de contenu multimédia vers les centres hospitaliers. Une solution de suivi de patient à distance doit être capable de fournir ces facilités qui imposent l’accès au haut débit. Dans une contribution répondant à cette préoccupation, nous suggérons un algorithme de réservation de ressources permettant de mieux gérer la qualité de service pour le contenu multimédia médical. / The aging of the population will probably catalyze the rise of chronic diseases and could intensify the need for personal assistance solutions. While researchers are focusing on information and communication technologies to provide responses to these public health problems, the number of connected objects is experiencing a rapid expansion. Indeed, desired revolution of technologies for health, forprevention and disease treatment coincides with the development of the Internet of Things 2. Thus, technological innovations and medical progress, for making it possible to monitor pathologies, often chronic, anywhere need appropriate equipments. Also, remote and real-time patient monitoring applications would require more network resources. In this context, communication resources management/sharing, technologies and equipments compatibilities and aplication’s desired performances become significant challenges. In this thesis, we propose an architecture based on Cognitive Radio, for meeting the medical applications constraints. We also analyze and test three important criteria for emergency transmissions, using this architecture.Connectivity : Any solution for patients monitoring must have anywhere and anytime capabilities for care continuity needs. High availability of network services and quality offered are critical for patient telemonitoring. We propose in this context, a spectral prediction mechanism able to examine the occupation conditions of the frequency bands. The algorithm we propose, associated learning and Grey Model technique in order to deal with any network connection discontinuities.Interference management : Network equipments must be able to perceive or to analyze their environment and act according to the underlying constraints. The interest is to protect in our case, medical equipment which are very sensitive to noise. Patient monitoring becomes possible at home or in the hospital, for example, with an acceptable level of interference. We propose for this criterion evaluation, a Cognitive Radio Networks deployment model in a hospital area. We define examples of functions for dynamic adaptation of the communication parameters, depending on the nearby medical devices sensitivity.Transmission efficiency under multimedia content delivery : This criterion analizes the ability of the architecture to provide desired quality in multimedia content delivery for real-time assistance or diagnosis. Patient monitoring at home or an emergency event may require the transmission of image or audio content to the hospital center. The remote monitoring solution must be able to provide these facilities which require a broadband network. We suggest an algorithm for resource reservation that performs a better management of the quality of service for medical multimedia content. We combine this algorithm with a transmission parameters control methode for maintaining the QoS at an acceptable level.

Les progrès récents des technologies de communication et la prolifération de l'informatique sans fil et des dispositifs de communication, ont induit 'a une surcharge dans l'utilisation du spectre radio. Cependant, les expériences de la Commission Fédérale de Communication (FCC) ont révélé que l'utilisation du spectre varie entre 15% et 85%. Par conséquent, les réseaux radios cognitifs (Cognitive Radio Networks ou CRNs) sont proposés afin d'utiliser le spectre radio d'une manière opportuniste. Dans ce type de réseaux radios cognitifs, où les fréquences de transmission sont sélectionnées d'une manière opportuniste - également sont appelés réseaux Ad-Hoc à radios cognitives -, la fiabilité de la dissémination des données est difficile 'a réaliser. D'abord, en plus des défis déjà connus dans les environnements sans fils, la diversité dans le nombre de fréquences qu'un noeud à radio cognitif a droit d'utiliser ajoute un autre défi, en limitant l'accessibilité à ses noeuds voisins. Deuxièmement, les noeuds à radio cognitif (CR) doivent conquérir les ressources de fréquences résiduelles avec les noeuds à radio primaire (PR), tout en essayent de les exploiter d'une manière opportuniste. En outre, les noeuds CR ne devraient pas perturber la qualité de réception des noeuds PR durant leur communication, et ce en limitant les interférences entre les deux de noeuds. Par conséquent, une nouvelle méthode de sélection de fréquences est requise afin de réduire le nombre d'interférences nuisibles aux noeuds PR, et maximiser les chances de délivrance des messages aux voisins récepteurs des noeuds CR, et augmenter ainsi la fiabilité des données disséminées. Dans cette thèse nous proposons SURF, une nouvelle méthode distribuée de sélection de fréquences pour la dissémination fiable de données dans un réseau radio cognitif multi-sauts. SURF classifie les fréquences radio disponibles en fonction de l'occupation des fréquences des noeuds à radio primaire et le nombre de noeuds 'a radio cognitive utilisant ces fréquences. Les résultats de simulation obtenus par NS-2 confirment que SURF est une stratégie efficace dans la sélection des meilleures fréquences de diffusion de données, comparée aux autres approches liées. Nous avons aussi constaté que les stratégies de sélection de fréquences sont considérablement influencées par l'activité des noeuds 'a radio primaire. Dans la suite ce cette thèse, nous étudierons et analyserons l'impact des modèles d'activités des noeuds PR sur les différentes stratégies de sélection de fréquences à travers des simulations basées NS-2. Nous avons remarqué que l'activité intermittente de PR est le cas où les solutions intelligentes doivent opérées. C'est dans ce cas où SURF donne les meilleures résultats et la région ciblée se serve des opportunités de communication. Enfin, dans cette thèse, nous allons encore plus loin en vérifiant l'applicabilité et la faisabilité de SURF. Dans cette perspective, d'abord, nous proposons une architecture d'accès à internet basse sur la radio cognitive pour les réseaux partiellement endommagés. Nous discutons les détails architecturaux et le principe de fonctionnement de l'architecture proposée. Nous avons également passé en revue les enjeux et les défis de déploiement de cette nouvelle architecture. Deuxièmement, nous discutons l'applicabilité de SURF dans le contexte de l'agrégation de fréquences et à cet égard, nous discutons une stratégie d'interférence basée sur l'agrégation de fréquences pour les réseaux radios cognitifs.

One critical step to successfully integrate wireless data networks to the high-speed wired backbone is the design of network control policies that efficiently utilize resources to provide Quality of Service (QoS) to the users in the integrated networks. Such a design has remained a challenge since wireless networks are time-varying in nature, not only in terms of user/packet arrivals but also in terms of physical channel conditions and access opportunities. In this thesis, we study the stochastic control of time-varying networks to design efficient scheduling and resource allocation policies. In particular, in Chapter 3, we focus on a broad class of control policies that work based on a pick-and-compare principle for networks with time-varying channels. By trading the throughput for complexity and memory requirement, these policies require less complexity compared to the well-investigated throughput-optimal Generalized Maximum Weight Matching (GMWM) policy and also require only linear-memory storage with the number of data-flows. Through Lyapunov analysis tools, we characterize the stability region and delay performance of the studied policies and show how they vary in response to the channel variations. In Chapter 4, we go into further detail and consider the problem of network control from a new perspective through which we carefully incorporate the time-efficiency of underlying scheduling algorithms. Specifically, we develop a policy that dynamically adjusts the time given to the available scheduling algorithms according to queue-backlog and channel correlations. We study the resulting stability region of developed policy and show that the region is at least as large as the one for any static policy. Finally, motivated by the current under-utilization of wireless spectrum, in Chapter 5, we investigate the control of cognitive radio networks as a special example of networks that provide time-varying access opportunities. We assume that users dynamically join and leave the network and may have different utility functions, or could collaborate for a common purpose. We develop a policy that performs joint admission and resource control and works for any user load, either inside or outside the capacity region. Through Lyapunov Optimization techniques, we show that the developed policy can achieve a utility performance arbitrarily close to the optimality with a tradeoff in the average service delay of admitted users.

One critical step to successfully integrate wireless data networks to the high-speed wired backbone is the design of network control policies that efficiently utilize resources to provide Quality of Service (QoS) to the users in the integrated networks. Such a design has remained a challenge since wireless networks are time-varying in nature, not only in terms of user/packet arrivals but also in terms of physical channel conditions and access opportunities. In this thesis, we study the stochastic control of time-varying networks to design efficient scheduling and resource allocation policies. In particular, in Chapter 3, we focus on a broad class of control policies that work based on a pick-and-compare principle for networks with time-varying channels. By trading the throughput for complexity and memory requirement, these policies require less complexity compared to the well-investigated throughput-optimal Generalized Maximum Weight Matching (GMWM) policy and also require only linear-memory storage with the number of data-flows. Through Lyapunov analysis tools, we characterize the stability region and delay performance of the studied policies and show how they vary in response to the channel variations. In Chapter 4, we go into further detail and consider the problem of network control from a new perspective through which we carefully incorporate the time-efficiency of underlying scheduling algorithms. Specifically, we develop a policy that dynamically adjusts the time given to the available scheduling algorithms according to queue-backlog and channel correlations. We study the resulting stability region of developed policy and show that the region is at least as large as the one for any static policy. Finally, motivated by the current under-utilization of wireless spectrum, in Chapter 5, we investigate the control of cognitive radio networks as a special example of networks that provide time-varying access opportunities. We assume that users dynamically join and leave the network and may have different utility functions, or could collaborate for a common purpose. We develop a policy that performs joint admission and resource control and works for any user load, either inside or outside the capacity region. Through Lyapunov Optimization techniques, we show that the developed policy can achieve a utility performance arbitrarily close to the optimality with a tradeoff in the average service delay of admitted users.

This dissertation develops optimal algorithms for distributed detection and estimation in static and mobile sensor networks. In distributed detection or estimation scenarios in clustered wireless sensor networks, sensor motes observe their local environment, make decisions or quantize these observations into local estimates of finite length, and send/relay them to a Cluster-Head (CH). For event detection tasks that are subject to both measurement errors and communication errors, we develop an algorithm that combines a Maximum a Posteriori (MAP) approach for local and global decisions with low-complexity channel codes and processing algorithms. For event estimation tasks that are subject to measurement errors, quantization errors and communication errors, we develop an algorithm that uses dithered quantization and channel compensation to ensure that each mote's local estimate received by the CH is unbiased and then lets the CH fuse these estimates into a global one using a Best Linear Unbiased Estimator (BLUE). We then determine both the minimum energy required for the network to produce an estimate with a prescribed error variance and show how this energy must be allocated amongst the motes in the network. In mobile wireless sensor networks, the mobility model governing each node will affect the detection accuracy at the CH and the energy consumption to achieve this level of accuracy. Correlated Random Walks (CRWs) have been proposed as mobility models that accounts for time dependency, geographical restrictions and nonzero drift. Hence, the solution to the continuous-time, 1-D, finite state space CRW is provided and its statistical behavior is studied both analytically and numerically. The impact of the motion of sensor on the network's performance is also studied.

The popularity of ubiquitous Internet services has spurred the fast growth of wireless communications by launching data hungry multimedia applications to mobile devices. Powered by spectrum agile cognitive radios, the newly emerged cognitive radio networks (CRN) are proposed to provision the efficient spectrum reuse to improve spectrum utilization. Unlicensed users in CRN, or secondary users (SUs), access the temporarily idle channels in a secondary and opportunistic fashion while preventing harmful interference to licensed primary users (PUs). To effectively detect and exploit the spectrum access opportunities released from a wide spectrum, the heterogeneous wireless channel characteristics and the underlying prioritized spectrum reuse features need to be considered in the protocol design and resource management schemes in CRN, which plays a critical role in unlicensed spectrum sharing among multiple users. The purpose of this dissertation is to address the challenges of utilizing heterogeneous wireless channels in CRN by its intrinsic dynamic and diverse natures, and build the efficient, scalable and, more importantly, practical dynamic spectrum access mechanisms to enable the cost-effective transmissions for unlicensed users. Note that the spectrum access opportunities exhibit the diversity in the time/frequency/space domain, secondary transmission schemes typically follow three design principles including 1) utilizing local free channels within short transmission range, 2) cooperative and opportunistic transmissions, and 3) effectively coordinating transmissions in varying bandwidth. The entire research work in this dissertation casts a systematic view to address these principles in the design of the routing protocols, medium access control (MAC) protocols and radio resource management schemes in CRN. Specifically, as spectrum access opportunities usually have small spatial footprints, SUs only communicate with the nearby nodes in a small area. Thus, multi-hop transmissions in CRN are considered in this dissertation to enable the connections between any unlicensed users in the network. CRN typically consist of intermittent links of varying bandwidth so that the decision of routing is closely related with the spectrum sensing and sharing operations in the lower layers. An efficient opportunistic cognitive routing (OCR) scheme is proposed in which the forwarding decision at each hop is made by jointly considering physical characteristics of spectrum bands and diverse activities of PUs in each single band. Such discussion on spectrum aware routing continues coupled with the sensing selection and contention among multiple relay candidates in a multi-channel multi-hop scenario. An SU selects the next hop relay and the working channel based upon location information and channel usage statistics with instant link quality feedbacks. By evaluating the performance of the routing protocol and the joint channel and route selection algorithm with extensive simulations, we determine the optimal channel and relay combination with reduced searching complexity and improved spectrum utilization. Besides, we investigate the medium access control (MAC) protocol design in support of multimedia applications in CRN. To satisfy the quality of service (QoS) requirements of heterogeneous applications for SUs, such as voice, video, and data, channels are selected to probe for appropriate spectrum opportunities based on the characteristics and QoS demands of the traffic along with the statistics of channel usage patterns. We propose a QoS-aware MAC protocol for multi-channel single hop scenario where each single SU distributedly determines a set of channels for sensing and data transmission to satisfy QoS requirements. By analytical model and simulations, we determine the service differentiation parameters to provision multiple levels of QoS. We further extend our discussion of dynamic resource management to a more practical deployment case. We apply the experiences and skills learnt from cognitive radio study to cellular communications. In heterogeneous cellular networks, small cells are deployed in macrocells to enhance link quality, extend network coverage and offload traffic. As different cells focus on their own operation utilities, the optimization of the total system performance can be analogue to the game between PUs and SUs in CRN. However, there are unique challenges and operation features in such case. We first present challenging issues including interference management, network coordination, and interworking between cells in a tiered cellular infrastructure. We then propose an adaptive resource management framework to improve spectrum utilization and mitigate the co-channel interference between macrocells and small cells. A game-theory-based approach is introduced to handle power control issues under constrained control bandwidth and limited end user capability. The inter-cell interference is mitigated based upon orthogonal transmissions and strict protection for macrocell users. The research results in the dissertation can provide insightful lights on flexible network deployment and dynamic spectrum access for prioritized spectrum reuse in modern wireless systems. The protocols and algorithms developed in each topic, respectively, have shown practical and efficient solutions to build and optimize CRN.

The most significant challenge for next wireless generation is to work opportunistically on the spectrum without a fixed spectrum allocation. Cognitive Radio (CR) is the candidate technology to utilize spectrum white space, which requires the CR to change its operating channel as the white space moves. In a CR ad-hoc network, each node could tune to a different channel; as a result, it cannot communicate with other nodes. This different tuning is due to the difficulty of maintaining Common Control Channel (CCC) in opportunistic spectrum network, and keeping the nodes synchronized in ad-hoc network. The CR ad-hoc network requires a protocol to match tuning channels between ad-hoc nodes, namely, rendezvous channels. In this thesis, two distributed Medium Access Control (MAC) protocols are designed that provide proper rendezvous channel without CCC or synchronization. The Balanced Incomplete Block Design (BIBD) is used in both protocols to provide our protocols a method of rendezvous between CR ad-hoc nodes. In fact, the BIBD guarantees there is at least one common element between any two blocks. If the channels are assigned to the BIBD elements and the searching sequence to the BIBD block, there is a guarantee of a rendezvous at least in one channel for each searching sequence. The first protocol uses a single-BIBD sequence and a multi-channel sensing. Alternatively, the second protocol uses a multi-BIBD sequence and a single-channel sensing. The single-sequence protocol analysis is based on the discrete Markov Chain. At the same time, the sequence structure of the BIBD in a multi-sequence protocol is used to define the Maximum Time to Rendezvous (MTTR). The simulation results confirm that the protocols outperform other existing protocols with respect to Time to Rendezvous (TTR), channel utilization, and network throughput. In addition, both protocols fairly distribute the network load on channels, and share the channels fairly among network nodes. This thesis provides straight forward and efficiently distributed MAC protocols for the CR ad-hoc networks.

The most significant challenge for next wireless generation is to work opportunistically on the spectrum without a fixed spectrum allocation. Cognitive Radio (CR) is the candidate technology to utilize spectrum white space, which requires the CR to change its operating channel as the white space moves. In a CR ad-hoc network, each node could tune to a different channel; as a result, it cannot communicate with other nodes. This different tuning is due to the difficulty of maintaining Common Control Channel (CCC) in opportunistic spectrum network, and keeping the nodes synchronized in ad-hoc network. The CR ad-hoc network requires a protocol to match tuning channels between ad-hoc nodes, namely, rendezvous channels. In this thesis, two distributed Medium Access Control (MAC) protocols are designed that provide proper rendezvous channel without CCC or synchronization. The Balanced Incomplete Block Design (BIBD) is used in both protocols to provide our protocols a method of rendezvous between CR ad-hoc nodes. In fact, the BIBD guarantees there is at least one common element between any two blocks. If the channels are assigned to the BIBD elements and the searching sequence to the BIBD block, there is a guarantee of a rendezvous at least in one channel for each searching sequence. The first protocol uses a single-BIBD sequence and a multi-channel sensing. Alternatively, the second protocol uses a multi-BIBD sequence and a single-channel sensing. The single-sequence protocol analysis is based on the discrete Markov Chain. At the same time, the sequence structure of the BIBD in a multi-sequence protocol is used to define the Maximum Time to Rendezvous (MTTR). The simulation results confirm that the protocols outperform other existing protocols with respect to Time to Rendezvous (TTR), channel utilization, and network throughput. In addition, both protocols fairly distribute the network load on channels, and share the channels fairly among network nodes. This thesis provides straight forward and efficiently distributed MAC protocols for the CR ad-hoc networks.

The objective of the proposed research is to develop interference-aware resource management techniques for CR networks that opportunistically operate within the licensed primary networks spectrum and to investigate the application of such CR techniques to emerging wireless networks. In this thesis, we report on a set of laboratory experiments that we undertook to analyze the interference between the CR-based wireless regional-area network (WRAN) standard and the digital television (DTV) broadcasting system. We determined the tolerable levels of WRAN interference into DTV receivers and studied the effect of these interference levels on WRAN deployment. Based on the need for efficient utilization of the primary network spectrum, we propose efficient interference-aware radio resource allocation (RRA) techniques for orthogonal frequency-division multiple access (OFDMA) CR networks. These RRA techniques aim to maximize the CR network throughput and to keep the CR interference to the primary network at or below a predefined threshold, known as the "interference temperature" limit. Moreover, we propose a joint spectrum-sensing design and power control algorithm that lead to increased CR network throughput and efficient protection of the PUs from undue interference. Interference coordination (IC) is considered a key technique for capacity maximization in emerging heterogeneous wireless networks. We propose a CR-based IC and RRA algorithm for OFDMA femtocell deployments to achieve efficient spectrum utilization and maximum network throughput. CR is envisioned as a key enabling technology for future wireless networks; our novel CR techniques will provide other researchers useful tools to design such networks.

The thesis presents research on multiobjective optimization based on memetic computing and its applications in engineering. We have introduced a framework for adaptive multiobjective memetic optimization algorithms (AMMOA) with an information theoretic criterion for guiding the selection, clustering, and local refinements. A robust stopping criterion for AMMOA has also been introduced to solve non-linear and large-scale optimization problems. The framework has been implemented for different benchmark test problems with remarkable results. This thesis also presents two applications of these algorithms. First, an optimal image data hiding technique has been formulated as a multiobjective optimization problem with conflicting objectives. In particular, trade-off factors in designing an optimal image data hiding are investigated to maximize the quality of watermarked images and the robustness of watermark. With the fixed size of a logo watermark, there is a conflict between these two objectives, thus a multiobjective optimization problem is introduced. We propose to use a hybrid between general regression neural networks (GRNN) and the adaptive multiobjective memetic optimization algorithm (AMMOA) to solve this challenging problem. This novel image data hiding approach has been implemented for many different test natural images with remarkable robustness and transparency of the embedded logo watermark. We also introduce a perceptual measure based on the relative Rényi information spectrum to evaluate the quality of watermarked images. The second application is the problem of joint spectrum sensing and power control optimization for a multichannel, multiple-user cognitive radio network. We investigated trade-off factors in designing efficient spectrum sensing techniques to maximize the throughput and minimize the interference. To maximize the throughput of secondary users and minimize the interference to primary users, we propose a joint determination of the sensing and transmission parameters of the secondary users, such as sensing times, decision threshold vectors, and power allocation vectors. There is a conflict between these two objectives, thus a multiobjective optimization problem is used again in the form of AMMOA. This algorithm learns to find optimal spectrum sensing times, decision threshold vectors, and power allocation vectors to maximize the averaged opportunistic throughput and minimize the averaged interference to the cognitive radio network. / February 2016