Submitted by Irvana Coutinho (irvana@ufpa.br) on 2015-02-13T14:08:11Z No. of bitstreams: 2 license_rdf: 22974 bytes, checksum: 99c771d9f0b9c46790009b9874d49253 (MD5) Dissertacao_GreenMarkovAbordageml.pdf: 1014480 bytes, checksum: 78962cf42379d2207770c49f4976e8ef (MD5) / Approved for entry into archive by Ana Rosa Silva (arosa@ufpa.br) on 2015-02-19T12:02:56Z (GMT) No. of bitstreams: 2 license_rdf: 22974 bytes, checksum: 99c771d9f0b9c46790009b9874d49253 (MD5) Dissertacao_GreenMarkovAbordageml.pdf: 1014480 bytes, checksum: 78962cf42379d2207770c49f4976e8ef (MD5) / Made available in DSpace on 2015-02-19T12:02:56Z (GMT). No. of bitstreams: 2 license_rdf: 22974 bytes, checksum: 99c771d9f0b9c46790009b9874d49253 (MD5) Dissertacao_GreenMarkovAbordageml.pdf: 1014480 bytes, checksum: 78962cf42379d2207770c49f4976e8ef (MD5) Previous issue date: 2014 / CNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O uso da comunicação de voz e dados através de dispositivos móveis vem aumentando significativamente nos últimos anos. Tal expansão traz algumas dificuldades inerentes, tais como: ampliação constante de capacidade das redes e eficiência energética. Neste contexto, vem se consolidando o conceito de Green networks, que se concentra no esforço para economia de energia e redução de CO2. Neste sentido, este trabalho propõe validar um modelo de uma política baseado em processo markoviano de decisão, visando a otimizar o consumo de energia, QoS e QoE, na alocação de usuários em redes macrocell e femtocell. Para isso o modelo foi inserido no simulador NS-2, aliando a solução analítica markoviana à flexibilidade característica da simulação discreta. A partir dos resultados apresentados na simulação, a política obteve uma economia significativa no consumo energético, melhorando a eficiência energética em até 4%, além de melhorar a qualidade de serviço em relação às redes macrocell e femtocell, demonstrando-se eficaz, de modo a alterar diretamente as métricas de QoS e de QoE. / The use of voice and data communication via mobile devices has increased significantly in recent years. This expansion brings some difficulties such as: continuous expansion of network capacity and energy efficiency. In this context, has been consolidating the concept of Green Networks, which focuses on the effort to energy saving and CO2 reduction. Thus, this paper proposes validate a model of a policy based on Markov decision process to optimize energy consumption, QoS and QoE, in the allocation of macrocell and femtocell users in networks. For this the model was inserted into the network simulator NS-2, combining the Markov analytical solution to the characteristic flexibility of discrete simulation. From the results presented in the simulation, the policy obtained significant savings in energy consumption, improving energy efficiency by up to 4%, and improve the quality of service in relation to the macrocell and femtocell networks, demonstrating effective in order to directly change the metrics of QoS and QoE.

La préservation de l’environnement et des ressources naturelles pour les prochaines générations est aujourd’hui considérée comme un des axes les plus prioritaires dans presque tous les secteurs économiques. Le secteur des Technologies de l’Information et de la Communication est loin d’être épargné de cette tendance écologique. Nous considérons dans cette thèse la problématique de la conservation d’énergie dans le contexte technologique actuel caractérisé par: • La coexistence d’une multitude de technologies d’accès sans fil offrant un environnement riche et dynamique • Des terminaux mobiles multimodaux • Limitations persistantes des sources d’énergie sur les terminaux mobiles. Dans ce contexte très riche, les possibilités offertes aux usagers sont à double tranchant. D’un côté, elles peuvent très bien améliorer la QoS en offrant toujours la meilleure connectivité en fonction du contexte de l’utilisateur. D’un autre côté, et sans une bonne optimisation de la consommation d’énergie sur le terminal, la disponibilité de celui-ci peut vite diminuer et donc faire baisser la QoE à cause de l’énergie nécessaire pour gérer plusieurs interfaces radio en parallèle. Nous considérons essentiellement les liens entre les stations de base (ou les point d’accès) et les terminaux mobiles. Notre objectif étant d’analyser la consommation d’énergie sur ces liens pour ensuite proposer des contributions permettant de mieux la maitriser. Nous focalisons essentiellement sur l’exploitation des multiples interfaces et du multi-flux pour étudier, analyser et proposer des solutions dynamiques et adaptatives d’ordonnancement, de sélection et de gestion d’interfaces minimisant la consommation d’énergie / Since the last decades, environmental issues are becoming among the major concerns for most human activities, including the Information and Communication Technologies sector. This will surely influence upcoming networking technologies, architectures and usage practices. New approaches and methodologies are required in order to evaluate and to reduce the Carbon Footprint toward what is commonly denoted as Green Networks. Within the ICT sector, the main efforts are related to energy saving techniques. These efforts started in early stages within wireless technologies, mainly because of energy limitations on mobile devices such as mobile phones and wireless sensors. Additionally, because of health considerations, standardization bodies and government had set stringent policies and limits on electromagnetic radiation levels that can be emitted by radio stations. For these reasons, many academic and industrial research and development activities had led to a number of relatively energy efficient solutions. In this thesis, we consider energy efficiency in the context of Heterogeneous Wireless Access Networks. These are composed of multi-standards wireless network solutions, with non uniform topologies and cell sizes and Multi-Modal mobile terminals able to manage simultaneously different connections. The main contributions of our studies include the proposal of new optimization solutions regarding user association and scheduling techniques at both flow and packet levels for multi-homed mobile terminals. An overall context-based solution is also proposed in order to provide end-to-end energy efficient networking solutions

Adaptive antenna systems (AAS's) are traditionally of interest only in radar and sonar applications. However, since the onset of the explosive growth in demand for wireless communications during the 1990's, researchers are giving increasing attention to the use of AAS technology to overcome practical challenges in providing the service. The main benefit of the technology lies in its ability to exploit the spatial domain, on top of the temporal and frequency domains, to improve on transceiver performance. This thesis presents a unified study on two classes of preprocessing techniques for uniform circular arrays (UCA's). UCA's are of interest because of their natural ability to provide a full azimuth (i.e. 360') coverage found in typical scenarios for sensor array applications, such as radar, sonar and wireless communications. The two classes of preprocessing techniques studied are the Davies transformation and the interpolated array transformations. These techniques yield a mathematically more convenient form - the Vandermonde form - for the array steering vector via a linear transformation. The Vandermonde form is useful for different applications such as direction-of-arrival (DOA) estimation and optimum or minimum variance distortionless response (MVDR) beamforming in correlated signal environment and beampattem synthesis. A novel interpolated array transformation is proposed to overcome limitations in the existing interpolated array transformations. A disadvantage of the two classes of preprocessing techniques for UCA's with omnidirectional elements is the lack of robustness in the transformed array steering vector to array imperfections under certain conditions. In order to mitigate the robustness problem, optimisation problems are formulated to modify the transformation matrices. / Suitable optimisation techniques are then applied to obtain more robust transformations. The improved transformations are shown to improve robustness but at the cost of larger transformation errors. The benefits of the robustification procedure are most apparent in DOA estimation. In addition to the algorithm level studies, the thesis also investigates the use of AAS technology with respect to two different third generation (3G) mobile communications systems: Enhanced Data rates for Global Evolution (EDGE) and Wideband Code Division Multiple Access (WCDMA). EDGE, or more generally GSM/EDGE Radio Access Network (GERAN), is the evolution of the widely successful GSM system to provide 3G mobile services in the existing radio spectrum. It builds on the TDMA technology of GSM and relies on improved coding and higher order modulation schemes to provide packet-based services at high data rates. WCDMA, on the other hand, is based on CDMA technology and is specially designed and streamlined for 3G mobile services. For WCDMA, a single-user approach to DOA estimation which utilises the user spreading code and the pulse-shaped chip waveform is proposed. It is shown that the proposed approach produces promising performance improvements. The studies with EDGE are concerned with the evaluation of a simple AAS at the system and link levels. / Results from, the system and link level simulations are presented to demonstrate the effectiveness of AAS technology in the new mobile communications system. Finally, it is noted that the WCDMA and EDGE link level simulations employ the newly developed COST259 directional channel model, which is capable of producing accurate channel realisations of macrocell environments for the evaluation of AAS's.