The main objective of the present thesis focuses on the integration of the in-wheel electric motors into the conception and control of road vehicles. The present thesis is the subject of the grant 186-654 (2010-2013) between the Laboratory of Signals and Systems (L2S-CNRS) and the French Institute of Petrol and New Energies (IFPEN). The thesis work has originally started from a vehicular electrification project, equipped with in-wheel electric motors at the rear axle, to obtain a full electric urban use and a standard extra-urban use with energy recovery at the rear axle in braking phases. The standard internal combustion engines have the disadvantage that complex estimation techniques are necessary to compute the instantaneous engine torque. At the same time, the actuators that control the braking system have some delays due to the hydraulic and mechanical circuits. These aspects represent the primary motivation for the introduction and study of the integration of the electric motor as unique propelling source for the vehicle. The advantages brought by the use of the electric motor are revealed and new techniques of control are set up to maximize its novelty. Control laws are constructed starting from the key feature of the electric motor, which is the fact that the torque transmitted at the wheel can be measured, depending on the current that passes through the motor. Another important feature of the electric motor is its response time, the independent control, as well as the fact that it can produce negative torques, in generator mode, to help decelerate the vehicle and store energy at the same time. Therefore, the novelty of the present work is that the in-wheel electric motor is considered to be the only control actuator signal in acceleration and deceleration phases, simplifying the architecture of the design of the vehicle and of the control laws. The control laws are focused on simplicity and rapidity in order to generate the torques which are transmitted at the wheels. To compute the adequate torques, estimation strategies are set up to produce reliable maximum friction estimation. Function of this maximum adherence available at the contact between the road and the tires, an adequate torque will be computed in order to achieve a stable wheel behavior in acceleration as well as in deceleration phases. The critical issue that was studied in this work was the non-linearity of the tire-road interaction characteristics and its complexity to estimate when it varies. The estimation strategy will have to detect all changes in the road-surface adherence and the computed control law should maintain the stability of the wheel even when the maximum friction changes. Perturbations and noise are also treated in order to test the robustness of the proposed estimation and control approaches.

Road accidents represent a serious social problem and are one of the leading causes of human death and disability on a global scale. To reduce the risk and severity of a road accident, a variety of new safety applications can be realized through wireless communications among vehicles driving nearby each other, or among vehicles and especially deployed road side units (RSUs), a technology known as a vehicular ad hoc network (VANET). Most of the VANET-enabled safety applications are based on broadcasting of safety messages by vehicles or RSUs, either periodically or in case of an unexpected event, such as a hard brake or dangerous road condition detection. Each broadcast safety message should be successfully delivered to the surrounding vehicles and RSUs without any excess delay, which is one of the main functions of a medium access control (MAC) protocol proposed for VANETs. This thesis presents VeMAC, a new multichannel time division multiple access (TDMA) protocol specifically designed to support the high priority safety applications in a VANET scenario. The ability of the VeMAC protocol to deliver periodic and event-driven safety messages in VANETs is demonstrated by a detailed delivery delay analysis, including queueing and service delays, for both types of safety messages. As well, computer simulations are conducted by using MATLAB, the network simulator ns-2, and the microscopic vehicle traffic simulator VISSIM, in order to evaluate the performance of the VeMAC protocol, in comparison with the IEEE 802.11p standard and the ADHOC MAC protocol (another TDMA protocol proposed for ad hoc networks). A real city scenario is simulated and different performance metrics are evaluated, including the network goodput, protocol overhead, channel utilization, protocol fairness, probability of a transmission collision, and safety message delivery delay. It is shown that the VeMAC protocol considerably outperforms the existing MAC schemes, which have significant limitations in supporting VANET safety applications. In addition to enhancing road safety, in-vehicle Internet access is one of the main applications of VANETs, which aims at providing the vehicle passengers with a low-cost access to the Internet via on-road gateways. This thesis presents a new strategy for deploying Internet gateways on the roads, in order to minimize the total cost of gateway deployment, while ensuring that a vehicle can connect to an Internet gateway (using multihop communications) with a probability greater than a specified threshold. This cost minimization problem is formulated by using binary integer programming, and applied for optimal gateway placement in a real city scenario. To the best of our knowledge, no previous strategy for gateway deployment has considered the probability of multihop connectivity among the vehicles and the deployed gateways. In order to allow a vehicle to discover the existence of an Internet gateway and to communicate with the gateway via multihops, a novel data packet routing scheme is proposed based on the VeMAC protocol. The performance of this cross-layer design is evaluated for a multichannel VANET in a highway scenario, mainly in terms of the end-to-end packet delivery delay. The packet queueing at each relay vehicle is considered in the end-to-end delay analysis, and numerical results are presented to study the effect of various parameters, such as the vehicle density and the packet arrival rate, on the performance metrics. The proposed VeMAC protocol is a promising candidate for MAC in VANETs, which can realize many advanced safety applications to enhance the public safety standards and improve the safety level of drivers/passengers and pedestrians on roads. On the other hand, the proposed gateway placement strategy and packet routing scheme represent a strong step toward providing reliable and ubiquitous in-vehicle Internet connectivity.

In this effort, we consider the performance of a dedicated short range communication (DSRC) system for inter-vehicle communications (IVC). The DSRC standard employs convolutional codes for forward error correction (FEC). The performance of the DSRC system is evaluated in three different channels with convolutional codes, regular low density parity check (LDPC) codes and quasi-cyclic (QC) LDPC codes. In addition, we compare the complexity of these codes. It is shown that LDPC and QC-LDPC codes provide a significant improvement in performance compared to convolutional codes. / Graduate

The number concentration and size distribution of particles in Brisbane have been studied extensively by the researchers at The International Laboratory for Air Quality and Health, Queensland University of Technology (Morawska et al., 1998, 1999a, 1999b). However, the comprehensive studies of chemical compositions of atmospheric particles, especially with regard to the two main classes of pollutants (polycyclic aromatic hydrocarbons and trace elements), that are usually of environmental and health interest, have not been fully undertaken. Therefore, this thesis presents detailed information on polycyclic aromatic hydrocarbons (PAHs) and elemental compositions of vehicle exhausts and of urban air in Brisbane. The levels of polycyclic aromatic hydrocarbons (PAHs) and elements in three of Brisbane's urban sites (Queensland University of Technology, Woolloongabba and ANZ stadium sites) were measured. The most common PAHs found in all sites were naphthalene, phenanthrene, anthracene, fluoranthene, pyrene and chrysene while Al, Cd, Co, Cr, Cu, Fe, Mn, Mo, Si, Sn, Sr and Zn were the most common elements detected in the total suspended particles and fine particle (PM2.5). With the aid of multivariate analysis techniques, several outcomes were obtained. For example: -- Major human activities such as vehicular and industrial sources were the most contributing pollution sources in Brisbane. However, these two sources have different influential strength on the compositions of the polycyclic aromatic hydrocarbons and trace inorganic elements found in the urban air. -- Woolloongabba bus platform was the most polluted site on the basis of the elemental and PAH compositions in its air samples while QUT site was the worst polluted site in terms of PM2.5 elemental contents. These results demonstrated that the impact of traffic related pollutants on Brisbane's urban air is significant. This led to the investigations of the direct emissions of pollutants from exhaust vehicular source in the second part of this research work. The exhaust studies included the investigations of PAHs, trace inorganic elements and particles. At the time of the study, the majority of vehicles in Brisbane used low sulfur diesel (LSD) fuel or unleaded petrol (ULP). However, the importance of vehicles using ultra low sulfur diesel (ULSD) and liquefied petroleum gas (LPG) is constantly growing. Therefore, the exhaust emission studies on chassis dynamometer from heavy duty non-catalyst-equipped buses powered by LSD and ULSD with 500 ppm and 50 ppm sulfur contents respectively as well as passenger cars powered by ULP and LPG were explored. The outcomes of such studies are summarized as follows: -- Naphthalene, acenaphthene, acenaphthylene, anthracene, phenanthrene, fluorene, fluoranthene and pyrene were frequently emitted by the buses powered by LSD and ULSD. However, buses powered by ULSD emitted 91% less PAHs than those powered by LSD. On the other hand, Mg, Ca, Cr, Fe, Cu, Zn, Ti, Ni, Pb, Be, P, Se, Ti and Ge were found in measurable quantities in the exhaust of the buses. The emissions of the elements were found to be strongly influenced by the engine driving conditions of the buses and fuel parameters such as sulfur content, fuel density and cetane index. -- Naphthalene, fluorene, phenanthrene, anthracene, pyrene, chrysene, benzo(a)anthracene and benzo(b)fluoranthene were predominantly emitted by ULP and LPG cars. On the average, the total emission factors of PAHs from LPG cars were generally lower than those of ULP cars, but given the large variations in the emission factors of cars powered by the same type of fuel, differences in the emission factors from both car types were statistically insignificant. In general, platinum group elements and many other elements were found in the exhausts of cars powered by both fuels. Emissions of inorganic elements from the cars were dependent on the type and the mileage of the cars. For example, ULP cars generally emitted higher levels of Cu, Mg, Al and Zn while LPG cars emitted higher level of V. In addition, cars with higher mileages were associated with higher emissions of the major elements (Zn, Al, Fe, V and Cu). -- Buses powered by ULSD usually emitted fewer particles, which were generally 31% to 59% lower than those emitted by LSD powered buses. Similarly, cars powered by LPG emitted less particles from those powered by ULP fuel. However, more nanoparticles (those with aerodynamic diameters of less than 50 nm) were emitted by LPG powered cars than their ULP counterparts. Health effect assessment of the exhaust PAHs was evaluated in terms of benzo(a)pyrene toxicity equivalent (BAPeq). The potential toxicities of PAHs emitted by ULSD powered buses were generally lower than those emitted by their LSD counterparts. A similar trend with lower emissions of PAHs from LPG cars than from ULP cars was observed when otherwise identical passenger cars were powered by LPG and ULP fuels. In summary, this thesis has shown that the majority of airborne particles found around Brisbane have anthropogenic origins, particularly vehicle emissions, and that fuel or lubricant formulations and engine operating conditions play important roles in the physical and chemical characteristics of pollutants emitted by vehicles. The implications of these results on worldwide strategies to reduce the environmental and health effects of particles emitted by motor vehicles were discussed. In this regard, direct emission measurements from vehicles powered by LSD, ULSD, ULP and LPG unveiled the relative environmental benefits associated with the use of ULSD in place of LSD to power diesel engines, and of LPG in place of ULP to power passenger cars.

No description available.

[pt] O presente trabalho apresenta um conceito para um sistema de evasão de colisões, simulado usando modelos 3D de três veículos diferentes implementados em MATLAB. Dois destes veículos foram parametrizados com dados genéricos, caracterizando automóveis de médio e grande porte. Em seguida, utilizados para realização de simulações iniciais e demonstração de conceitos. O terceiro conjunto de dados foi construído com informações do Apollo N, um veículo super esportivo. Estes diferentes conjuntos de dados foram utilizados para avaliar a capacidade do controlador de trabalhar com veículos de diferentes portes e dinâmicas de direção. A abordagem para acionar o sistema baseia-se no cálculo do tempo para a colisão (TTC; timeto- collision). O conceito foi adotado para detectar situações onde o motorista não é capaz de evitar um acidente. Depois de ser acionado, o sistema deve decidir qual manobra é a mais apropriada, dadas as condições de aderência da pista e o risco associado. O primeiro objetivo deste trabalho é desenvolver um sistema autônomo de frenagem que deve ser capaz de avaliar o risco de uma possível colisão e decidir se o condutor é capaz de evitá-la. Uma vez que o motorista não tenha tempo suficiente para reagir, o sistema deve acionar os freios automaticamente a fim de evitar um possível acidente. Além disso, o veículo possui um sistema anti-travamento (ABS), desenvolvido usando controle Fuzzy. O desempenho do controlador ABS foi avaliado em simulações usando os conjuntos de dados e testado em um veículo em escala. Em casos mais críticos, quando há baixa aderência, o veículo não é capaz de frear em uma distância razoável. Levando-se em consideração tal situação, um controle autônomo de esterçamento também foi desenvolvido, visando a possibilidade de uma manobra alternativa de evasão. Este segundo sistema foi avaliado em simulações utilizando veículos com características subesterçantes e sobreesterçantes. Os resultados mostraram que o controle de esterçamento foi capaz de realizar manobras evasivas produzindo valores razoáveis de acelerações laterais, em veículos com diferentes dinâmicas de direção. / [en] This work presents a concept for a collision avoidance system simulated using 3D-models of three different vehicles implemented in MATLAB. Two of the vehicle data sets were built with generic information, used to characterize mid-size and full-size vehicles. These standard vehicles were used in initial simulations and for demonstration of some concepts. The third data set was built with information from the Apollo N, a super sportive car. These different data sets were used to evaluate the controller s capacity to work with a range of vehicles, with different sizes and driving characteristics. The approach for triggering the system is based on the time-to-colision (TTC) estimation. This concept was adopted to recognize when the driver is not able to avoid an accident. After being triggered, the system must decide which maneuver is the most appropriate for the given friction and risk conditions. The first goal of this work is to develop an autonomous braking system which evaluates the risk of a possible collision and decides if the driver is able to avoid it. Once the driver has not enough time to react, the system must trigger the brakes automatically in order to avoid the accident. The vehicle is equipped with an embedded Anti-lock Brake System (ABS) developed using Fuzzy control. The ABS controller s performance was evaluated in simulations using the data sets and tested in a scaled vehicle. In more critical cases, when there is low friction, the vehicle is not able to brake in a reasonable distance. Considering this situation, an autonomous steering control was implemented in order to make an alternative avoidance maneuver. This second system was evaluated in simulations using vehicles with understeering and oversteering characteristics. The results pointed out that the autonomous steering control was able to perform avoidance maneuvers in a reasonable range of lateral accelerations, in vehicles with different driving tendencies. / [de] Die vorliegende Arbeit prasentiert ein Konzept fur ein Kollisionsvermeidungssystem. Dieses wird anhand von drei verschiedenen 3DFahrzeugmodellen mit Hilfe von MATLAB simuliert. Zwei der FahrzeugDatensatze basieren auf generischen Informationen, die jeweils ein Automobil der Mittelklasse und der Oberklasse reprasentieren. Diese Standardfahrzeuge wurden fur anfangliche Simulationen und zur Demonstration einiger Konzepte verwendet. Das dritte Fahrzeugmodell wurde mit Hilfe der Daten des Sportwagens Apollo N aufgebaut. Durch die Verwendung der verschiedenen Datensatze soll die Funktionsfahigkeit der Regelung auch bei verschiedenen Fahrzeugtypen mit unterschiedlichen Dimensionen und Fahreigenschaften uberpruft werden.Die Grundlage zum Auslosen des Systems ist die Abschatzung der Zeit bis zur Kollision (TTC; time-to-collision). Dieses Konzept wurde aufgegriffen, um zu entscheiden, wann der Fahrer nicht mehr in der Lage ist einen Unfall zu vermeiden. Nachdem das System ausgelost wird muss dieses anhand der Traktionsverhaltnisse und Gefahrensituation entscheiden, welches Manover am besten geeignet ist. Das erste Teilziel ist die Entwicklung eines autonomen Bremssystems, welches eine bevorstehende Kollision erkennen muss und entscheidet ob der Fahrer die Kollision eigenstandig vermeiden kann. Sobald der Fahrer nicht mehr genug Zeit hat selbst zu reagieren, muss das System die Bremsen automatisch betatigen um den Unfall zu vermeiden. Hierzu ist das Fahrzeug mit einem Antiblockiersystem (ABS) ausgestattet. Dieses wurde mit Hilfe eines Fuzzy-Kontrollers realisiert. Die Funktionstuchtigkeit der ABS-Regelung wurde mit Simulationen und anhand eines realen, skalierten Fahrzeugmodells getestet. In kritischen Situationen, kann es aufgrund der Traktionsverhaltnisse vorkommen, dass das Fahrzeug nicht mehr in der Lage ist innerhalb einer ausreichenden Strecke zum Stehen zu kommen. Um fur solche Situationen ein alternatives Ausweichmanöver anwenden zu konnen, wurde ein automatischer Lenkeingriff implementiert. Dieses System wurde anhand von Simulationen an Fahrzeugmodellen mit Ubersteuernden und Untersteuernden Eigenschaften uberprüft. Die Ergebnisse zeigten, dass die automatische Lenkeingriff-Regelung in der Lage war auch bei Fahrzeugen mit unterschiedlichen Fahreigenschaften Ausweichmanöver unter Einhaltung angemessener Querbeschleunigungen durchzufuhren.

Les réseaux véhiculaires constituent une catégorie de réseaux sans fil mobiles à part entière et présentent l'originalité de permettre aux véhicules de communiquer les uns avec les autres mais aussi avec l'infrastructure quand elle existe. L'apparition des réseaux véhiculaires s'est accompagnée de l'apparition d'une myriade et variété d'applications potentielles allant de la sécurité à la gestion du trafic routier en passant par les applications de divertissement et de confort des usagers de la route. Ces applications ont suscité beaucoup d'intérêt de la part des chercheurs, des constructeurs des automobiles et des opérateurs des télécommunications. Les applications d'information et de divertissement, pour lesquelles une grande quantité de contenus peut exister, exigent que les contenus engendrés soient propagés au travers des véhicules et/ou de l'infrastructure jusqu'à atteindre les utilisateurs intéressés tout en respectant les durées de vie potentiellement limitées des contenus. La dissémination de contenus pour ce type d'applications reste un défi majeur en raison de plusieurs facteurs tels que la présence de beaucoup de contenus, la connectivité très intermittente mais encore les intérêts potentiellement hétérogènes des utilisateurs. C'est à cette thématique que nous sommes intéressés dans cette thèse; Tout d'abord, nous nous proposons une nouvelle métrique qui calcule l'utilité apportée aux utilisateurs. Elle permet de mesurer leur satisfaction par rapport aux contenus reçus. Nous la jugeons nécessaire pour évaluer les performances d'une approche de dissémination pour les applications de confort par opposition à des applications de sécurité routière. Dans un deuxième temps, nous nous concentrons sur le développement d'un nouveau protocole de dissémination, appelé I-PICK, et d'une solution de sélection des nœuds relais, appelé I-SEND, pour disséminer les contenus d'information et de divertissement en tenant compte des préférences des utilisateurs par rapport aux contenus reçus. Notre proposition est fondée sur l'échange de messages périodiques permettant l'estimation des durées de contacts et la connaissance des préférences des utilisateurs. Ces informations sont ensuite utilisées, dans un premier temps, pour effectuer un ordonnancement efficace des contenus lors de la dissémination puis choisir les relais permettant de maximiser l'utilité des utilisateurs par rapport aux contenus reçus dans un environnement caractérisé par des faibles durées de communication. Au travers de simulations nous confirmons l'efficacité de notre approche. Pour conforter le fonctionnement de nos mécanismes, nous avons implanté dans un environnement réel nos propositions I-PICK et I-SEND. Au travers d'un scénario simple, nous avons mis en évidence des risques liés à l'hétérogénéité des machines ou bien encore la difficulté du paramétrage des temporisations. Ces premiers résultats positifs montrent l'intérêt de notre technique et ouvre des pistes d'amélioration. Notre dernière contribution concerne des mécanismes de réduction du trafic cellulaire à l'aide des communications opportunistes entre les véhicules. Quand un contenu est disponible auprès d'un serveur de contenus accessible par le réseau cellulaire, il est nécessaire de proposer une méthode efficace de sélection des sources initiales qui seront choisies pour télécharger puis disséminer les contenus. Nous optons pour une solution qui pourrait reposer sur technologie SDN ainsi que sur des communications opportunistes et qui permet de choisir les sources en tant que nœuds pouvant produire un maximum d'utilité en propageant les contenus. / Vehicular networks are a class of mobile wireless networks and have the originality of enabling vehicles to communicate with each other and also with the infrastructure when it exists. The advent of vehicular networks has been accompanied by the emergence of a myriad and a variety of potential applications that are not only restricted to road safety but span from traffic management to entertainment and comfort applications. These applications have received much interest from researchers, automobile manufacturers, and telecommunications operators. Information and entertainment applications, where a large amount of content can exist, require the dissemination of the generated content through vehicles and/or infrastructure until reaching interested users while respecting the potential limited lifetime of content. The content dissemination for this type of applications remains a major challenge due to several factors such as the presence of a large amount of content, an intermittent connectivity, and the heterogeneous interests of users. In this thesis we turn our attention to this topic. First, we propose a new metric that allows to measure the users satisfaction with respect to the received content. This metric is used to evaluate the performance of the dissemination schemes. Second, we focus on the development of a new dissemination protocol, named I-PICK, and a forwarder selection solution, named I-SEND, to disseminate information and entertaining content in vehicular networks while ensuring maximum satisfaction of the users preferences with respect to the received content. Our proposal is based on the exchange of periodic messages enabling the estimation of contact durations and the knowledge of user preferences. First of all, this information is used to conduct an efficient content scheduling during the dissemination, and then to select relay nodes for maximizing the utility to users. We confirm the efficiency of our approach through simulations. To strengthen the functioning of our mechanisms, we implemented our proposals I-PICK and I-SEND in a real environment. Through a simple scenario, we have highlighted the risks associated with the heterogeneity of the machines and even the difficulty of setting timers. These first positive results show the interest of our solutions and raise new questions to be addressed. Our last contribution concerns cellular traffic offloading schemes by exploiting vehicular opportunistic communications. We study the seed-vehicles selection problem as the first step toward bootstrapping cellular traffic offloading for content dissemination in vehicular networks. We propose a solution that takes advantage from the presence of the SDN technology as well as vehicular opportunistic communications to select as seeds the nodes that can produce maximum utility when propagating the content.

Made available in DSpace on 2016-06-02T19:06:08Z (GMT). No. of bitstreams: 1 5463.pdf: 18006027 bytes, checksum: 047b84b38eb03b475dacbf51b7bf50b1 (MD5) Previous issue date: 2013-07-31 / Financiadora de Estudos e Projetos / Vehicular ad-hoc networks (VANETs) are networks capable of establishing communications between vehicles and road-side units. VANETs could be employed in data transmission applications. However, due to vehicle mobility, VANETs present intermittent connectivity, making message transmission a challenging task. Due to the lack of an end-to-end connectivity, messages are forwarded from vehicle to vehicle and stored when it is not possible to retransmit. Additionally, in order to improve delivery probability, messages are replicated and disseminated over the network. However, message replication may cause high network overhead and resource usage. As result, considerable research e_ort has been devoted to develop algorithms for speci_c scenarios: low, moderate and high connectivity. Nevertheless, algorithms projected for scenarios with a speci_c connectivity lack the ability to adapt to situations with zones presenting diferent node density. This lack of adaptation may negatively a_ect the performance in application such as data transmission in cities. This masters project proposes develops a method to automatically adapt message replication routing algorithms to diferent node density scenarios. The proposed method is composed of three phases. The first phase collects data from message retransmission events using a standard routing algorithms. The second phase consists in training a decision tree classifier based on the collected data. Finally, in the third phase the trained classifier is used to determine whether a message should be retransmitted or not based on the local node density. Therefore, the proposed method allows routing algorithms to query the trained classifier to decide if a message should be retransmitted. The proposed method was evaluated with real movement traces in order to improve Spray and Wait and Epidemic routing algorithms. Results indicate that the proposed method may contribute to performance enhancement. / As VANETs são redes de veículos com capacidade de estabelecer comunicações sem fio entre veículos e com equipamentos nas estradas. Estas redes poderiam ser usadas para a transferência de dados de diversas aplicações. No entanto, devido á mobilidade dos veículos, as VANETs apresentam conectividade intermitente entre os nós, dificultando a transmissão de mensagens. Ante a impossibilidade de ter conectividade de fim a fim, as mensagens são encaminhadas progressivamente de veículo em veículo, e armazenadas quando não houver a possibilidade de retransmitir. Adicionalmente, para incrementar a probabilidade de entrega, as mensagens são replicadas e disseminadas pela rede. Não obstante, a replicação de mensagens pode gerar alta sobrecarga de rede e alto consumo de recursos. Por causa disto, projetaram-se algoritmos para cenários específicos de: baixa, moderada e alta conectividade. Estes algoritmos, quando aplicados em ambientes de zonas de diferente densidade de nós,como cidades, podem diminuir o seu desempenho pela falta da capacidade de se adaptar a diferentes condições de conectividade. Contudo, neste trabalho foi desenvolvido um método para adaptar o comportamento dos algoritmos de roteamento por replicação de mensagens a diferentes situações de conectividade segundo a densidade das zonas onde se movimentam os nós retransmissores. O método consiste em três fases. Na primeira, são coletados os dados dos eventos de repasse de mensagens utilizando o algoritmo de roteamento padrão. Na segunda fase, utilizam-se os dados coletados para treinar um classificador baseado em _arvores de decisão. Na _ultima fase, o classificador é então empregado para determinar se uma situação de repasse de mensagem _e favorável segundo a densidade de nós. Desta forma, os algoritmos de roteamento podem decidir se repassar ou não uma mensagem com o suporte do classificador. Esta abordagem foi avaliada com traces de movimentos reais, para aprimorar o desempenho dos algoritmos de roteamento Spray and Wait e Epidemic. Os resultados dos experimentos realizados revelam que esta abordagem pode contribuir para o aprimoramento do desempenho.

Le degré de pénétration du marché des appareils intelligents tels que les smartphones et les tablettes avec les technologies de communication embarquées comme le WiFi, 3G et LTE a explosé en moins d’une décennie. En complément de cette tendance technologique, les applications des réseaux sociaux ont virtuellement connectées une grande partie de la popula- tion, en génerant une demande croissante de trafic de données vers et depuis l’infrastructure de communication. Les communications pervasives ont aussi acquis une importance dans l’industrie automobile. L’émergence d’une gamme impressionnante d’appareils intelligents dans les véhicules a permis des services tels que : l’assistante au conducteur, l’infotainment, le suivi à distance du véhicule, et la connectivité aux réseaux sociaux même en déplacement La demande exponentielle de connectivité a encore défié les fournisseurs de services de télé- communications pour répondre aux attentes des utilisateurs.L’objectif de cette thèse est de modéliser et comprendre la mobilité dynamique des utilisateurs à grande vitesse et leurs effets sur les architectures de réseau sans fil. Compte tenu de l’ importance du développement de notre étude sur une représentation réal- iste de la mobilité des véhicules, nous étudions tout d’abord les approches les plus populaires pour la génération de trafic routier synthétique et discutons les caractéristiques des ensem- bles de données accessibles au public qui décrivent des mobilités véhiculaires. En utilisant l’information des déplacements de la population dans une région métropolitaine, les données détaillées du réseau routier et les modèles réalistes des conduites microscopiques, nous pro- posons un jeu de données de mobilité véhiculaire original qui redéfinit l’état de l’art et qui replie la circulation routière de façon réaliste dans le temps et dans l’espace. Nous étudions ensuite l’impact des dynamiques des mobilité du point de vue de la couverture cellulaire en présence d’un déploiement réel des stations de base. En outre, en examinant les effets de la mobilité des véhicules sur les réseaux autonomes, nous voyons des possibilités pour les futurs paradigmes de réseaux hétérogènes. Motivés par l’évolution dynamique dans le temps, de la mobilité des véhicules observée dans notre jeux de données, nous proposons également une approche en ligne pour prédire les flux de trafic macroscopiques. Nous analysons les paramètres affectant la prédiction de la mobilité en milieu urbain. Nous dévoilons quand et où la gestion des ressources réseaux est la plus cruciale pour accueillir le trafic généré par les utilisateurs à bord. Ces études révèlent de multiples opportunités de gestion intelligente des transports, soit pour construire de nouvelles routes, soit pour l’installation de bornes de recharge électriques, ou pour la conception de systèmes de feux de circulation intelligents, contribuant ainsi à la planification urbaine. / The market penetration of smart devices like smartphones and tablets with embedded communication technologies like WiFi, 3G and LTE has exploded in less than a decade. Complementing this technological trend, social networking applications have virtually connected a large portion of the population generating an ever-growing data traffic demand on the communication infrastructure. Pervasive communications have gained significance in the automobile industry as well, with the emergence of an impressive range of in-vehicle smart devices enabling driver assistance, infotainment, over-the-air vehicle monitoring, and even social connectivity on the move. This surge in the demand for connectivity has further challenged telecommunication service providers to meet the expectations of high-speed network users. The goal of this thesis is to model and understand the mobility dynamics of high-speed users and their effect on wireless network architectures. Given the importance of developing our study on a realistic representation of vehicular mobility, we first survey the most popular approaches for the generation of synthetic road traffic and discuss the features of publicly available vehicular mobility datasets. Using original travel demand information of the population of a metropolitan area, detailed road network data and realistic microscopic driving models, we propose a novel state-of-art vehicular mobility dataset that closely mimics the real-world road traffic dynamics in both time and space. We then study the impact of such mobility dynamics from the perspective of wireless cellular network architecture in presence of a real-world base station deployment. In addition, by discussing the effects of vehicular mobility on autonomous network architecture, we hint at the opportunities for future heterogenous network paradigms. Motivated by the time-evolving mobility dynamics observed in our original dataset, we also propose an online approach to predict near-future macroscopic traffic flows. We analyze the parameters affecting the mobility prediction in an urban environment and unveil when and where network resource management is more crucial to accommodate the traffic generated by users onboard. Such studies unveil multiple opportunities in transportation management either for building new roads, installing electric charging points, or for designing intelligent traffic light systems, thereby contributing to urban planning.

This licentiate thesis work investigates two medium access control (MAC) methods, when used in traffic safety applications over vehicular ad hoc networks (VANETs). The MAC methods are carrier sense multiple access (CSMA), as specified by the leading standard for VANETs IEEE 802.11p, and self-organizing time-division multiple access (STDMA) as used by the leading standard for transponders on ships. All vehicles in traffic safety applications periodically broadcast cooperative awareness messages (CAMs). The CAM based data traffic implies requirements on a predictable, fair and scalable medium access mechanism. The investigated performance measures are channel access delay, number of consecutive packet drops and the distance between concurrently transmitting nodes. Performance is evaluated by computer simulations of a highway scenario in which all vehicles broadcast CAMs with different update rates and packet lengths. The obtained results show that nodes in a CSMA system can experience unbounded channel access delays and further that there is a significant difference between the best case and worst case channel access delay that a node could experience. In addition, with CSMA there is a very high probability that several concurrently transmitting nodes are located close to each other. This occurs when nodes start their listening periods at the same time or when nodes choose the same backoff value, which results in nodes starting to transmit at the same time instant. The CSMA algorithm is therefore both unpredictable and unfair besides the fact that it scales badly for broadcasted CAMs. STDMA, on the other hand, will always grant channel access for all packets before a predetermined time, regardless of the number of competing nodes. Therefore, the STDMA algorithm is predictable and fair. STDMA, using parameter settings that have been adapted to the vehicular environment, is shown to outperform CSMA when considering the performance measure distance between concurrently transmitting nodes. In CSMA the distance between concurrent transmissions is random, whereas STDMA uses the side information from the CAMs to properly schedule concurrent transmissions in space. The price paid for the superior performance of STDMA is the required network synchronization through a global navigation satellite system, e.g., GPS. That aside since STDMA was shown to be scalable, predictable and fair; it is an excellent candidate for use in VANETs when complex communication requirements from traffic safety applications should be met.