O uso de equipamentos sem fio em sistemas de monitoramento e controle de processos industriais vem aumentando gradativamente. Os elevados custos de cabeamento incentivam os gerentes de fábricas para que considerem o uso de sistemas industriais sem fio, uma vez que estes promovem grandes reduções nos custos de instalação finais. Entretanto, a robustez do enlace de rádio frequência e os requerimentos de comunicação em tempo real são frequentemente citados como obstáculos reais para o uso de tecnologias sem fio na indústria. À frente das demais organizações, a HART Foundation lançou em 2007 o WirelessHART, sendo este o primeiro padrão aberto de comunicação sem fio especificamente desenvolvido para ambientes industriais. Apesar de ser um protocolo seguro e de propiciar comunicação em tempo real, o estudo aprofundado do WirelessHART revela diversas possibilidades de melhorias. Nesse protocolo, o gerenciamento da rede é feito de forma centralizada, o que garante maior simplicidade e controle sobre o escalonamento, roteamento e a segurança das mensagens. Por outro lado, grandes latências, especialmente nos processos de conexão e desconexão de nós, impossibilitam o uso de dispositivos móveis ou intermitentes. A concentração do gerenciamento em um só elemento também leva à fragilidade do sistema: na falha do gerenciador central, toda rede fica inoperante. Neste trabalho, é apresentada uma arquitetura de rede WirelessHART inovadora, que utiliza coprocessadores gerenciadores distribuídos para possibilitar a descentralização do gerenciamento e consequentemente as suas vantagens. O sistema desenvolvido permite diferentes configurações descentralizadas, parciais ou totais, e com mínimas modificações no protocolo original. A arquitetura de rede proposta mantém os dispositivos de campo com suas características originais, de baixo custo e consumo, além de permitir a coexistência com equipamentos certificados. A metodologia para a descentralização é apresentada e o estudo de caso realizado permite a avaliação da proposta. A originalidade deste trabalho é verificada através do estudo do estado da arte. Os resultados obtidos de um estudo de caso demonstram o grande potencial da proposta, com reduções significativas nos tempos de agregação e manutenção da rede. A proposta de descentralização possibilita configurações e modos operacionais antes impossíveis no WirelessHART, tais como o uso de dispositivos móveis ou intermitentes e o emprego de técnicas de handover. / The use of wireless devices in industrial systems for process monitoring and control has been gradually increasing. The high costs of cabling encourage plant managers to consider the use of such industrial wireless systems, since they promote large reductions in final costs. However, the link robustness and the real-time communication requirements are frequently cited as real obstacles to the use of wireless technologies in the industry. Ahead of other organizations, the HART Foundation launched the in 2007WirelessHART, which is the first open standard for wireless communications developed specifically for industrial environments. WirelessHART is a reliable and secure protocol and provide real-time communications but the thorough study of this protocol reveals several possibile improvements. In this protocol, network management is done centrally, ensuring simplicity and control over the scheduling, routing and messaging security. Moreover, large latency, especially in the connection and disconnection processes of field devices, preclude the use of mobile or intermittent devices. The concentration of one element in management also leads to the system fragility: if the central manager fails the entire network fail. In this work, a innovative WirelessHART network achitecture is presented and implemented by means of hardware-based distributed network manager coprocessors to enable decentralization and its blessings. The developed system allows different decentralized, partial or total architectures with minimal changes to the original protocol. A hardware-based solution maintains field devices with its low cost and consumption features and allow coexistence with certified equipment. The decentralization methodology along with its originality is presented and case study allows the evaluation of the proposal. The results demonstrate the architecture has great potential with significant reductions in join and maintenance times. The proposed decentralization enables configurations and operating modes previously impossible in WirelessHART such as the use of mobile or intermittent devices and handover techniques.

A melhora na eficiência do uso do espectro de radiofrequência é fundamental para permitir um desempenho adequado dos diversos sistemas sem fio cuja complexidade e requerimentos aumentam a cada dia. O panorama atual de atribuição de canais é estático, tolerando assim, que o espectro de radiofrequência seja usado de forma desequilibrada, gerando com isso, problemas de coexistência em algumas faixas e subutilização de recursos em outras. Para contornar esse problema, tem sido proposta a ideia de introduzir algoritmos de cognição nos dispositivos sem fio, a fim de permitir um modelo de alocação dinâmico adicional. Neste, os usuários secundários equipados com rádios cognitivos podem utilizar de forma dinâmica os canais subutilizados de usuários primários. Um usuário primário tem prioridade de uso do canal como consequência da atribuição estática, porém, o uso do canal por um usuário secundário qualquer é oportunista e limitando ao tempo de inatividade do usuário primário em um determinado local. As redes de sensores sem fio trabalham em uma banda concorrida e são sistemas que podem melhorar seu desempenho utilizando um mecanismo de acesso dinâmico aos canais, possibilitando o aproveitamento dos períodos de inatividade de usuários primários ou aumentando sua capacidade de coexistência na banda de operação atual. Segundo a literatura pesquisada são vários os desafios existentes para conseguir um método distribuído de acesso dinâmico aos canais que considere as restrições de trocas de sinalizações, consumo de energia e complexidade dos dispositivos de uma rede de sensores sem fio. Neste contexto, propõe-se uma metodologia de sensoriamento e acesso dinâmico aos canais para uma rede de sensores sem fio considerando a simplicidade dos dispositivos. Como fatores de inovação, optou-se pela definição de uma política de sensoriamento por clusters que permite realizar uma aprendizagem cooperativa por reforço da situação dos canais de operação. Além disso, a definição de um mecanismo de acesso dinâmico aos canais fundamentado no padrão IEEE 802.15.4 permite comunicação e coordenação distribuída de forma assíncrona. O funcionamento da metodologia proposta é avaliado e comparado usando simulações e experimentos mediante um estudo de caso específico. As comparações são realizadas com métodos de seleção de canal: fixa, cega ou baseada em recompensas por acesso. Os resultados mostram a eficiência no acesso dinâmico aos canais com aumentos na taxa de entrega de mensagens e na capacidade de coexistir com as redes primárias. / The efficiency improvement of the use of radiofrequency spectrum is fundamental to allow more complex and more optimal wireless systems. The current channel allocation is static. It tolerates unbalanced use of the radiofrequency spectrum generating coexistence problems in some bands and underutilization of resources in other bands. The introduction of cognitive algorithms into wireless devices has been proposed to overcome that problem, in order to allow an additional dynamic allocation model. In this, the secondary users equipped with cognitive radios will be able to use dynamically the underutilized channels of primary users. A primary user has channel usage priority related to the static allocation, on the other hand, the use of the channel by any secondary user is opportunistic and limited to the inactivity time of the primary user in a specific place. Wireless sensor networks work in a competitive band. These systems can improve their performance using a dynamic access to the channels and consequently to enable the utilization of inactivity periods of primary users or to increase the coexistence capability at their current operation band. According to the researched literature, several challenges exist to find a distributed method for dynamic access to the channels considering the restrictions on control signaling, energy consumption and computational complexity of wireless sensor network devices. In this context, a methodology of sensing and dynamic access to the channels in a wireless sensor network considering the restrictions of the devices is proposed. As innovation, we opted for the definition of a sensing policy by clusters that allows the cooperative reinforcement learning of the situation of channels. In addition, a mechanism for dynamic access to the channels based on the IEEE 802.15.4 standard is defined to allow asynchronous and distributed coordination. The behavior of the proposed methodology is evaluated and compared using simulations and experiments through a specific case study. The comparisons are performed with channel selection methods: fixed, blind and access based. The results show good efficiency in the dynamic allocation of the channels, increasing the message delivery rate and the coexistence capability.

The wireless sensor network (WSN) is increasingly used in many areas nowadays. It can be applied to provide the solutions to environmental problems, help increasing security and safety systems, and make the detection of the problems more efficient, e.g. the earthquake or tidal wave, which will harmful to humans. The WNS is durable and resistant to all types of terrain and climate, but while the WSN system is more and more widespread, one of the obstacles hindering the growth of this technology and the demand for WSN applications is the limited battery lifespan. Consequently, there is a significant requirement for techniques for prolonging the battery’s lifespan. Therefore, one potential solution is to use alternative energy sources combined with the sensor nodes in WSN, specifically energy harvesting from existing environmental sources. This research project reviews the characteristics of each kind of energy harvesting, understanding the various energy sources (solar energy, vibration energy and wind power), including wireless power transfer (WPT) by using electromagnetic (EM) radiation energy transfer or RF radio-frequency emission and magnetic coupled energy transfer. They are adopted for extending node’s life in the WSN, based on published information. Then it compares these diverse alternative energy methods and identifies for the most suitable energy harvesting method for application to wireless sensor nodes in order to prolong the lifespan of the battery. The major findings from the researcher include that wireless power transfer energy harvesting (WPT) using the magnetic field is the most appropriate tool for extending the lifespan of the WSN system. In addition, the author also designed an experiment to test this alternative energy, achieving by modelling the wireless power transfer with four coils. From the experimental results, it can be seen that the WPT technique using energy harvesting with magnetic inductive source can be applied to prolong the lifespan of the WSN system.

Les insectes ravageurs constituent depuis longtemps une menace sérieuse pour l’agriculture dans plusieurs régions du monde. En particulier, les criquets pèlerins (Schistocerca Gregaria) sont les plus redoutés à cause des dégâts colossaux qu’ils peuvent infliger à une large variété de cultures. Ces derniers, dans leur milieu de reproduction et un ensemble de paramètres climatiques, peuvent être assimilés à un système biophysique. Un système biophysique associe plusieurs composants caractérisés par leur complexité propre et celle de leurs interactions. Par exemple la pluie apporte de l’humidité et le développement du couvert végétal qui sont critiques dans la dynamique de population du criquet pèlerin. Ces systèmes biophysiques peuvent être contrôlés par divers moyens, dont un réseau de surveillance couplant des capteurs sans fil. L’intégration des systèmes biophysiques et des réseaux d’observation peut être un des faits majeurs de cette décennie. Elle permet un échange entre les processus cycliques physiques et les systèmes d’information. Ces derniers permettent de collecter et de traiter les données de terrain et peuvent aussi agir sur le système physique par le biais de boucles rétro-actives. Cette intégration a été possible grâce aux avancées technologiques notées dans les domaines de la micro électronique et de la transmission sans fil. On la trouve dans des domaines émergents tels que l’agriculture fine. Cette thèse est consacrée à la modélisation cellulaire et à la simulation de systèmes physiques. Pour cela, un échantillonnage spatial et temporel d’une zone de reproduction des criquets pèlerins a été réalisé à l’aide d’outils spécifiques avec différentes résolutions, dans le but de générer les systèmes cellulaires. Ces systèmes cellulaires contiennent des informations telles que le vent, la température, l’humidité relative et la pluviométrie. L’échantillonnage spatial permet une représentation de migration et l’échantillonnage temporel permet de suivre l’évolution locale des individus dans une cellule.Les systèmes cellulaires sont transcrits en réseaux de processus communicant et évoluent de manière synchrone afin de reproduire et de simuler le phénomène de migration intercellulaire et le cycle de vie du criquet pèlerin dans une cellule. La construction de ces systèmes a été explorée en profondeur, en faisant varier leur contexte géographique, les critères de classification, la dimension des cellules, la connectivité entre les processus, les comportements collectifs. L’exploration a porté sur la génération de codes parallèles et les performances à l’exécution, essentiellement pour le cas des processus légers.Ce travail a abouti à la production de deux simulateurs paramétriques, le premier est destiné à la dynamique de population des criquets pèlerins en vue d’évaluer l’état de la situation acridienne par un couplage entre les réseaux de surveillance synchrones et les systèmes biophysiques. Le second permet de planifier le déploiement de réseaux de capteurs sans fil dans une zone afin de déterminer l’emplacement des capteurs. / Insect pests have long been a serious threat to agriculture in many parts of the world. In particular, desert locusts (Schistocerca Gregaria) are the most feared because of the colossal damage they can inflict on a wide variety of crops.The latter, in their breeding environment and a set of climatic parameters, can be assimilated to a biophysical system. A biophysical system associates several components characterized by their own complexity and that of their interactions. For example, rain brings moisture and canopy development that is critical in Desert Locust population dynamics. These biophysical systems can be controlled by various means, including a monitoring network coupling wireless sensors. The integration of biophysical systems and observation networks can be one of the major events of this decade. It allows an exchange between physical cyclical processes and information systems. These can collect and process field data and can also act on the physical system through feedback loops. This integration has been made possible thanks to technological advances noted in the fields of microelectronics and wireless transmission. It is found in emerging fields such as fine agriculture. This thesis is devoted to cellular modeling and simulation of physical systems. For this, a spatial and temporal sampling of a Desert Locust breeding area was carried out using specific tools with different resolutions, in order to generate the cellular systems. These cellular systems contain information such as wind, temperature, relative humidity and rainfall. Spatial sampling allows a migration representation and temporal sampling can track the local evolution of individuals in a cell.Cellular systems are transcribed into communicating process networks and evolve synchronously to reproduce and simulate the phenomenon of intercellular migration and the life cycle of the Desert Locust in a cell. The construction of these systems has been explored in depth, making vary their geographical context, the criteria for classification, the cells, connectivity between processes, collective behaviors.Exploration has focused on the generation of parallel codes and performance at the execution, mainly for the case of lite processes.This work has led to the production of two parametric simulators, the first of which is aimed at Desert Locust population dynamics with a view to assessing the state of the locust situation through coupling between synchronous monitoring networks and biophysical systems. The second allows you to plan the deployment of wireless sensor networks in an area to determine the location of the sensors.

Les pannes sont la règle et non l'exception dans les réseaux de capteurs sans fil. Un nœud capteur est fragile et il peut échouer en raison de l'épuisement de la batterie ou de la destruction par un événement externe. En outre, le nœud peut capter et transmettre des valeurs incorrectes en raison de l'influence de l'environnement sur son fonctionnement. Les liens sont également vulnérables et leur panne peut provoquer un partitionnement du réseau et un changement dans la topologie du réseau, ce qui conduit à une perte ou à un retard des données. Dans le cas où les nœuds sont portés par des objets mobiles, ils peuvent être mis hors de portée de la communication. Les réseaux de capteurs sont également sujets à des attaques malveillantes, telles que le déni de service, l'injection de paquets défectueux, entraînant un comportement inattendu du système et ainsi de suite. En plus de ces défaillances prédéfinies (c'est-à-dire avec des types et symptômes connus), les réseaux de capteurs présentent aussi des défaillances silencieuses qui ne sont pas connues à l'avance, et qui sont très liées au système. En revanche, les applications de RCSF, en particulier les applications de sécurité critiques, telles que la détection d'incendie ou les systèmes d'alarme, nécessitent un fonctionnement continu et fiable du système. Cependant, la garantie d'un fonctionnement correct d'un système pendant l'exécution est une tâche difficile. Cela est dû aux nombreux types de pannes que l'on peut rencontrer dans un tel système vulnérable et non fiable. Une approche holistique de la gestion des fautes qui aborde tous les types de fautes n'existe pas. En effet, les travaux existants se focalisent sur certains états d'incohérence du système. La raison en est simple : la consommation d'énergie augmente en fonction du nombre d'éléments à surveiller, de la quantité d'informations à collecter et parfois à échanger. Dans cette thèse, nous proposons un «Framework » global pour la gestion des fautes dans un réseau de capteurs. Ce framework, appelé « IFTF », fournit une vision complète de l'état du système avec la possibilité de diagnostiquer des phénomènes anormaux. IFTF détecte les anomalies au niveau des données, diagnostique les défaillances de réseau, détecte les défaillances d'applications, et identifie les zones affectées du réseau. Ces objectifs sont atteints grâce à la combinaison efficace d'un service de diagnostic réseau (surveillance au niveau des composants), un service de test d'applications (surveillance au niveau du système) et un système de validation des données. Les deux premiers services résident sur chaque nœud du réseau et le système de validation des données réside sur chaque chef de groupe. Grâce à IFTF, les opérations de maintenance et de reconfiguration seront plus efficaces, menant à un système WSN (Wireless Sensor Network) plus fiable. Du point de vue conception, IFTF fournit de nombreux paramètres ajustables qui le rendent approprié aux divers types d'applications. Les résultats de simulation montrent que la solution présentée est efficace en termes de coût mémoire et d'énergie. En effet, le système de validation des données n'induit pas un surcoût de communication. De plus, le fonctionnement des deux services test et diagnostic augmente la consommation d'énergie de 4% en moyenne, par rapport au fonctionnement du service de diagnostic uniquement. / Sensor faults are the rule and not the exception in every Wireless Sensor Network (WSN) deployment. Sensor nodes are fragile, and they may fail due to depletion of batteries or destruction by an external event. In addition, nodes may capture and communicate incorrect readings because of environmental influence on their sensing components. Links are also failure-prone, causing network partitions and dynamic changes in network topology, leading to delays in data communications. Links may fail when permanently or temporarily blocked by an external or environmental condition. Packets may be corrupted due to the erroneous nature of communications. When nodes are embedded or carried by mobile objects, nodes can be taken out of the range of communications. WSNs are also prone to malicious attacks, such as denial of service, injection of faulty packets, leading to unexpected behavior of the system and so on. In addition to these predefined faults or failures (i.e., with known types and symptoms), many times the sensor networks exhibits silent failures that are unknown beforehand and highly system-related. Applications over WSNs, in particular safety critical applications, such as fire detection or burglar alarm systems, require continuous and reliable operation of the system. However, validating that a WSN system will function correctly at run time is a hard problem. This is due to the numerous faults that can be encountered in the resource constrained nature of sensor platforms together with the unreliability of the wireless links networks. A holistic fault management approach that addresses all fault issues does not exist. Existing work most likely misses some potential causes of system failures. The reason is simple : the more elements to monitor, the more information to be collected and sometimes to be exchanged, then the more the energy consumption becomes higher. In this thesis, we propose an Integrated Fault Tolerance Framework (IFTF) that provides a complete picture of the system health with possibility to zoom in on the fault reasons of abnormal phenomena. IFTF detects data anomalies, diagnoses network failures, detects application level failures, identifies affected areas of the network and may determine the root causes of application malfunctioning. These goals are achieved efficiently through combining a network diagnosis service (component/element level monitoring) with an application testing service (system level monitoring) and a data validation system. The first two services reside on each node in the network and the data validation system resides on each cluster head. Thanks to IFTF, the maintenance and reconfiguration operations will be more efficient leading to a more dependable WSN. From the design view, IFTF offers to the application many tunable parameters that make it suitable for various application needs. Simulation results show that the presented solution is efficient both in terms of memory use and power consumption. Data validation system does not incur power consumption (communication overhead). Using testing service combined to diagnosis service incurs a 4 %, on average, increase in power consumption compared to using solely network diagnosis solutions.

Les contraintes présentes dans les réseaux de capteurs impliquent l'introduction de techniques d'optimisation à différents niveaux de conception : du matériel au logiciel et dans la pile de communication. En effet, le déploiement des réseaux de capteurs, à faible consommation énergétique, exige une conception conjointe du matériel et du logiciel adaptée à l'application visée. Étant donné la nature évènementielle et multitâche des applications dans les réseaux de capteurs, nous pourrions penser à rajouter différentes unités de traitement qui coopèrent pour gérer les évènements et les tâches de manière optimale. Ainsi, la complexité des tâches accomplies par le processeur principal peut être réduite, ce qui contribue à atteindre l'efficacité énergétique. Dans cette thèse nous étudions un ensemble de protocoles qui facilitent l'implémentation des smart radios. L'idée principale des smart radios est l'introduction de l'intelligence dans la puce radio de manière à ce qu'elle soit capable de prendre des décisions ainsi que d'exécuter plusieurs tâches de manière autonome et sans l'intervention du processeur principal. Cette dernière sera responsable du bootstrap du réseau et, après qu'un état stable est atteint, le processeur peut rester inactif la plupart du temps, alors que la puce radio continue à fournir un ensemble de services. Le protocole proposé est appelé Wake on Idle et il fournit la maintenance de voisinage intégrée avec une méthode d'accès au canal. Ces services sont basés sur des transmissions analogiques qui sont codées dans le temps. De cette manière, dès que le réseau entre dans l'état stable (c.à.d. la topologie est formée et les noeuds sont associés et synchronisés), le traitement numérique de trames n'est pas nécessaire. Puisque Wake on Idle est basé sur des informations de bas niveau, il peut être facilement intégré dans la puce radio et fonctionner comme un coprocesseur qui fournit des services de haut niveau au processeur principal, comme la maintenance du voisinage et l'accès au canal. Grâce à une analyse théorique et une implémentation préliminaire, nous démontrons la faisabilité du protocole et nous montrons plusieurs caractéristiques intéressantes qui aident à atteindre l'efficacité énergétique et de bonnes performances. Ensuite, nous exploitons la signalisation analogique afin d'optimiser le duty-cycle des protocoles d'accès au canal existants. Nous proposons également un mécanisme appelé Sleep on Idle qui est basé sur l'échange de signaux analogiques ou busy tones. Sleep on Idle peut être intégré dans la radio et il peut décider quand le processeur doit être réveillé. Enfin, nous avons intégré le mécanisme de notification dans le standard IEEE802.15.4 et nous avons évalué ce mécanisme par des simulations et expérimentations. Les résultats montrent un gain important en termes de consommation en énergie et de réactivité du réseau. / The constraints of Wireless Sensor Networks scenarios require the introduction of optimization techniques at different design levels: from the hardware to the software and communication protocol stack. In fact, the design of energy efficient WSNs involves an appropriate hardware/software co-design oriented to the concerned application. Given the event driven and multitasking nature of WSNs applications, one could think of adding different processing units that cooperate to manage events and tasks in an optimal way. Then, the complexity of tasks performed by the main processing unit can be reduced and energy efficiency can be achieved. In this PhD thesis we study protocols that leverage the implementation of smart radios. The idea of smart radios is introducing intelligence into the radio chip; in this way, it will be able to take decisions and perform several tasks in an autonomous way and without any intervention of the main processing unit. The processing unit will be in charge of bootstrapping the network and, after a stable state is reached, it can remain inactive most of the time while the radio chip provides a given set of services. The proposed protocol is called Wake on Idle and it provides integrated neighborhood maintenance and low duty-cycle medium access control. These services are provided based on analog transmissions that are time encoded; then, as soon as the network enters the stable state (i.e. the topology is formed and nodes are associated and synchronized) digital processing of frames is not needed. Since it relies on low-level information, Wake on Idle can be easily implemented on hardware and integrated into the radio chip; then, it works as a coprocessor that provides high-level services (i.e. neighborhood maintenance and medium access) to the main processing unit. Through theoretical analysis and a preliminary implementation we demonstrate the feasibility of the protocol and we show several interesting characteristics that help achieving energy efficiency and good performance. Then, we further exploit analog signaling to optimize duty cycle of existing medium access control protocols. We propose a mechanism called Sleep on Idle and it is based on the exchange of analog busy tones. Sleep on Idle can also be integrated into the smart radio to take decisions about whether the main processing unit has to be woken up. We apply the decision mechanism to the slotted ieee802.15.4 standard and validate it through simulations and experimentations. The results show an important gain in terms of energy consumption and network reactivity.

Ce travail de recherche vise à concevoir un système de vision à faible consommation d'énergie pour les réseaux de capteurs sans fil. L'imageur en question doit respecter les contraintes spécifiques des applications multimédias pour les réseaux de capteurs de vision sans fil. En effet, de par sa nature, une application multimédia impose un traitement intensif au niveau du noeud et un nombre considérable de paquets à échanger à travers le lien radio, et par conséquent beaucoup d'énergie à consommer. Une solution évidente pour diminuer la quantité de données transmise, et donc la durée de vie du réseau, est de compresser les images avant de les transmettre. Néanmoins, les contraintes strictes des noeuds du réseau rendent inefficace en pratique l'exécution des algorithmes de compression standards (JPEG, JPEG2000, MJPEG, MPEG, H264, etc.). Le système de vision à concevoir doit donc intégrer des techniques de compression d'image à la fois efficaces et à faible complexité. Une attention particulière doit être prise en compte en vue de satisfaire au mieux le compromis "Consommation énergétique - Qualité de Service (QoS)". / This research aims to develop a vision system with low energy consumption for Wireless Sensor Networks (WSNs). The imager in question must meet the specific requirements of multimedia applications for Wireless Vision Sensor Networks. Indeed, a multimedia application requires intensive computation at the node and a considerable number of packets to be exchanged through the transceiver, and therefore consumes a lot of energy. An obvious solution to reduce the amount of transmitted data is to compress the images before sending them over WSN nodes. However, the severe constraints of nodes make ineffective in practice the implementation of standard compression algorithms (JPEG, JPEG2000, MJPEG, MPEG, H264, etc.). Desired vision system must integrate image compression techniques that are both effective and with low-complexity. Particular attention should be taken into consideration in order to best satisfy the compromise "Energy Consumption - Quality of Service (QoS)".

A melhora na eficiência do uso do espectro de radiofrequência é fundamental para permitir um desempenho adequado dos diversos sistemas sem fio cuja complexidade e requerimentos aumentam a cada dia. O panorama atual de atribuição de canais é estático, tolerando assim, que o espectro de radiofrequência seja usado de forma desequilibrada, gerando com isso, problemas de coexistência em algumas faixas e subutilização de recursos em outras. Para contornar esse problema, tem sido proposta a ideia de introduzir algoritmos de cognição nos dispositivos sem fio, a fim de permitir um modelo de alocação dinâmico adicional. Neste, os usuários secundários equipados com rádios cognitivos podem utilizar de forma dinâmica os canais subutilizados de usuários primários. Um usuário primário tem prioridade de uso do canal como consequência da atribuição estática, porém, o uso do canal por um usuário secundário qualquer é oportunista e limitando ao tempo de inatividade do usuário primário em um determinado local. As redes de sensores sem fio trabalham em uma banda concorrida e são sistemas que podem melhorar seu desempenho utilizando um mecanismo de acesso dinâmico aos canais, possibilitando o aproveitamento dos períodos de inatividade de usuários primários ou aumentando sua capacidade de coexistência na banda de operação atual. Segundo a literatura pesquisada são vários os desafios existentes para conseguir um método distribuído de acesso dinâmico aos canais que considere as restrições de trocas de sinalizações, consumo de energia e complexidade dos dispositivos de uma rede de sensores sem fio. Neste contexto, propõe-se uma metodologia de sensoriamento e acesso dinâmico aos canais para uma rede de sensores sem fio considerando a simplicidade dos dispositivos. Como fatores de inovação, optou-se pela definição de uma política de sensoriamento por clusters que permite realizar uma aprendizagem cooperativa por reforço da situação dos canais de operação. Além disso, a definição de um mecanismo de acesso dinâmico aos canais fundamentado no padrão IEEE 802.15.4 permite comunicação e coordenação distribuída de forma assíncrona. O funcionamento da metodologia proposta é avaliado e comparado usando simulações e experimentos mediante um estudo de caso específico. As comparações são realizadas com métodos de seleção de canal: fixa, cega ou baseada em recompensas por acesso. Os resultados mostram a eficiência no acesso dinâmico aos canais com aumentos na taxa de entrega de mensagens e na capacidade de coexistir com as redes primárias. / The efficiency improvement of the use of radiofrequency spectrum is fundamental to allow more complex and more optimal wireless systems. The current channel allocation is static. It tolerates unbalanced use of the radiofrequency spectrum generating coexistence problems in some bands and underutilization of resources in other bands. The introduction of cognitive algorithms into wireless devices has been proposed to overcome that problem, in order to allow an additional dynamic allocation model. In this, the secondary users equipped with cognitive radios will be able to use dynamically the underutilized channels of primary users. A primary user has channel usage priority related to the static allocation, on the other hand, the use of the channel by any secondary user is opportunistic and limited to the inactivity time of the primary user in a specific place. Wireless sensor networks work in a competitive band. These systems can improve their performance using a dynamic access to the channels and consequently to enable the utilization of inactivity periods of primary users or to increase the coexistence capability at their current operation band. According to the researched literature, several challenges exist to find a distributed method for dynamic access to the channels considering the restrictions on control signaling, energy consumption and computational complexity of wireless sensor network devices. In this context, a methodology of sensing and dynamic access to the channels in a wireless sensor network considering the restrictions of the devices is proposed. As innovation, we opted for the definition of a sensing policy by clusters that allows the cooperative reinforcement learning of the situation of channels. In addition, a mechanism for dynamic access to the channels based on the IEEE 802.15.4 standard is defined to allow asynchronous and distributed coordination. The behavior of the proposed methodology is evaluated and compared using simulations and experiments through a specific case study. The comparisons are performed with channel selection methods: fixed, blind and access based. The results show good efficiency in the dynamic allocation of the channels, increasing the message delivery rate and the coexistence capability.

Redes de sensores sem fio (RSSFs) heterogêneas podem combinar nós estáticos e nós móveis. Os nós móveis podem ainda conter um hardware mais sofisticado quando comparado aos nós estáticos. Veículos aéreos não tripulados (VANTs) podem conferir mobilidade ao nó sensor aumentando a flexibilidade da RSSF onde ele está inserido. Tanto os VANTs quanto outros n´os sensores comuns podem conter uma arquitetura de hardware reconfiguraável, como por exemplo um FPGA, e com isso adquirir um poder computacional diferenciado. RSSFs propiciam um grande e interessante espectro de aplicações possíveis, tais como vigilância aérea, suporte `a segurança pública entre outros. As RSSFs podem ser configuradas através do uso de agentes móveis, que são capazes de migrar carregando as tarefas que serão executadas nos n´os. Neste cenário, este trabalho descreve uma arquitetura de agentes reconfiguráveis para redes de sensores sem fio. Os agentes são capazes de serem executados como um agente puramente em software, ou também como um agente em hardware, de- pendendo do ambiente de execução disponível e do design escolhido. A arquitetura proposta para o agente reconfiguraável apresenta a transparência necessária ao agente para que o resto do sistema não perceba a natureza dos agentes que estão sendo executados na plataforma. Além disso, a arquitetura permite a migração dinâmica de agentes que reconfiguram o sistema também de uma maneira transparente. São mostrados exemplos de casos de uso que demostram a viabilidade de uso da arquitetura proposta e este trabalho ainda mostra a análise realizada sobre estas plataformas. / Heterogeneous wireless sensor networks (WSN) can combine static nodes and mobile nodes. Theses mobile nodes may contain a sophisticated hardware when compared to static nodes. Unmanned aircraft vehicles (UAVs) can confer mobility to the sensor node increasing the flexibility of the WSN to where it is inserted. UAVs as well as other common sensor nodes can have a reconfigurable hardware architecture, as, for example, an FPGA and with this achieve a differentiated computational power. WSNs enable a vast and interesting spectrum of possible applications, like aerial surveillance, public security support, among others. The WSNs can be configured by the use of mobile agents, which are capable of migrating among the nodes, carrying the tasks to be executed and that will be instantiated on a given node. In this scenario, this work describes an architecture of reconfigurable agents to wireless sensor networks. The agents can be implemented purely in software or as a hardware agent, depending on the available execution environment and on the chosen design. The proposed architecture presents the necessary transparency to the agent so that the rest of the system is not aware of the nature of the agents that are implemented on the platform. Furthermore, the architecture enables dynamic migration of agents that reconfigure the system in a transparent way as well. In this work, use cases examples that demonstrate the feasibility of using the proposed architecture are shown, as well as the analysis performed on these platforms.

O uso de equipamentos sem fio em sistemas de monitoramento e controle de processos industriais vem aumentando gradativamente. Os elevados custos de cabeamento incentivam os gerentes de fábricas para que considerem o uso de sistemas industriais sem fio, uma vez que estes promovem grandes reduções nos custos de instalação finais. Entretanto, a robustez do enlace de rádio frequência e os requerimentos de comunicação em tempo real são frequentemente citados como obstáculos reais para o uso de tecnologias sem fio na indústria. À frente das demais organizações, a HART Foundation lançou em 2007 o WirelessHART, sendo este o primeiro padrão aberto de comunicação sem fio especificamente desenvolvido para ambientes industriais. Apesar de ser um protocolo seguro e de propiciar comunicação em tempo real, o estudo aprofundado do WirelessHART revela diversas possibilidades de melhorias. Nesse protocolo, o gerenciamento da rede é feito de forma centralizada, o que garante maior simplicidade e controle sobre o escalonamento, roteamento e a segurança das mensagens. Por outro lado, grandes latências, especialmente nos processos de conexão e desconexão de nós, impossibilitam o uso de dispositivos móveis ou intermitentes. A concentração do gerenciamento em um só elemento também leva à fragilidade do sistema: na falha do gerenciador central, toda rede fica inoperante. Neste trabalho, é apresentada uma arquitetura de rede WirelessHART inovadora, que utiliza coprocessadores gerenciadores distribuídos para possibilitar a descentralização do gerenciamento e consequentemente as suas vantagens. O sistema desenvolvido permite diferentes configurações descentralizadas, parciais ou totais, e com mínimas modificações no protocolo original. A arquitetura de rede proposta mantém os dispositivos de campo com suas características originais, de baixo custo e consumo, além de permitir a coexistência com equipamentos certificados. A metodologia para a descentralização é apresentada e o estudo de caso realizado permite a avaliação da proposta. A originalidade deste trabalho é verificada através do estudo do estado da arte. Os resultados obtidos de um estudo de caso demonstram o grande potencial da proposta, com reduções significativas nos tempos de agregação e manutenção da rede. A proposta de descentralização possibilita configurações e modos operacionais antes impossíveis no WirelessHART, tais como o uso de dispositivos móveis ou intermitentes e o emprego de técnicas de handover. / The use of wireless devices in industrial systems for process monitoring and control has been gradually increasing. The high costs of cabling encourage plant managers to consider the use of such industrial wireless systems, since they promote large reductions in final costs. However, the link robustness and the real-time communication requirements are frequently cited as real obstacles to the use of wireless technologies in the industry. Ahead of other organizations, the HART Foundation launched the in 2007WirelessHART, which is the first open standard for wireless communications developed specifically for industrial environments. WirelessHART is a reliable and secure protocol and provide real-time communications but the thorough study of this protocol reveals several possibile improvements. In this protocol, network management is done centrally, ensuring simplicity and control over the scheduling, routing and messaging security. Moreover, large latency, especially in the connection and disconnection processes of field devices, preclude the use of mobile or intermittent devices. The concentration of one element in management also leads to the system fragility: if the central manager fails the entire network fail. In this work, a innovative WirelessHART network achitecture is presented and implemented by means of hardware-based distributed network manager coprocessors to enable decentralization and its blessings. The developed system allows different decentralized, partial or total architectures with minimal changes to the original protocol. A hardware-based solution maintains field devices with its low cost and consumption features and allow coexistence with certified equipment. The decentralization methodology along with its originality is presented and case study allows the evaluation of the proposal. The results demonstrate the architecture has great potential with significant reductions in join and maintenance times. The proposed decentralization enables configurations and operating modes previously impossible in WirelessHART such as the use of mobile or intermittent devices and handover techniques.