M. Tech. Electrical Engineering / It is widely believed that the implementation of multi-user multiple-input multiple- output (MU-MIMO) technology at the radio access portion of current and future wireless networks would positively impact on the performance of such networks. This however demands the design of efficient multi-user scheduling algorithms at the data link layer. One algorithm that is known to be throughput optimal is the opportunistic scheduler. This work takes a cross-layer approach in designing a scheduler that takes into account both the channel and queue states of users. We propose a scheduling algorithm referred to as the Maximum Throughput Scheduler (MTS). The MTS explicitly expresses network throughput as the utility function. Through simulations, we show that the MTS outperforms other schedulers in terms of network throughput.

Cognitive radio (CR) is a multi-disciplinary field that makes use of knowledge from a multitude of specialties such as antenna design, circuits, systems and digital signal processing among many others. CR has emerged as an area of interest over 20 years ago and in the years since has evolved to encompass both realizable theory and physical hardware. Key among the latter are reconfigurable, software defined radios and embedded sensors that incorporate flexible parameters, allowing a CR to operate in a wide variety of electromagnetic (EM) environments. The ideal cognitive radio would be capable of adapting to a changing EM environment without any specific knowledge or direction from the operator. This would require the radio itself to be aware of the EM environment and ideally, to sense the EM environment and act upon it in a semi-autonomous or autonomous way. While most research in this field has focused on the spectrum sensing aspects of the domain, development of the above-described "ideal CR" would require that the EM environment be characterized in domains such as angular, time and polarization among others. Signal dependent parameters can also be characterized such as bandwidth and modulation. The multi-dimensionality of the environment and the signals present within entail challenges with scalability and efficiency. This work focuses on the efficient sensing of signals in the angular domain also known as direction-of-arrival (DOA). There are a multitude of ways to find a signal's DOA. All require multiple antennas connected to a single or multiple radio nodes, antennas with patterns that gather energy in a particular direction, or multiple single antenna radios. The methods that utilize multiple antennas exploit the phase and/or amplitude relationships between the antennas themselves for a signal's DOA. The principal tradeoff between DOA methods typically converges to scan time vs. number of antenna elements. For many DOA architectures, this also means a scan time tradeoff with angular resolution as well. Since fast and accurate measurements are important for characterizing a quickly changing EM environment, sensing speed becomes a key requirement in designing a CR and associated sensing architecture. In this work, we present a DOA sensing architecture suitable for use in CR systems called the Direct Space to Information Converter (DSIC). Unlike current state-of-the art DOA methods, the DSIC breaks the tradeoff between scan time and the number of antenna elements needed for a given angular resolution when compared to other DOA and beamforming architectures. By randomly modulating the received signals in space, across multiple antenna elements and taking a few, compressed sensing (CS) measurements, the DSIC is able to angularly scan a wide field of view in an order of magnitude less time than other DOA methods. These CS measurements correspond to different random perturbations of the DSIC's antenna factor and can be quantized in as little as a single bit of resolution in the DSIC's phaseshifters/vector modulators. The DSIC is able to create multiple user-specified nulls in the antenna pattern to reduce the impact of strong known interferers while also simultaneously scanning the full field of view. Additionally, the designer has the option of performing simultaneous reception or nulling while sensing. If nulling, a few different methods are available each suitable for varying EM environments and potential use cases. We show in detail the multi-disciplinary process in designing a complete end-to-end hardware solution, selecting the parameters necessary to design the DSIC as well as test and characterize it. The benefits of the DSIC are discussed and compared to the current state-of the art with an emphasis on architectures suitable for use in interferer rich environments. We demonstrate that the energy usage of the DSIC is lower than comparable CR architectures by a large factor and scales much more favorably in terms of energy and physical complexity as the number of antenna elements increase. At the conclusion of this work we also discuss future areas of exploration in extending the DSIC's capability by incorporating an ability to sense the spectrum as well as the DOA of a signal.

Thesis (S.M.)--Joint Program in Oceanographic Engineering (Massachusetts Institute of Technology, Dept. of Ocean Engineering, and the and Woods Hole Oceanographic Institution), 2000. / Includes bibliographical references (p. 100-101). / A finite difference computer model is developed to simulate the exposure statistics of a radio frequency buoyant antenna as it is towed in a three-dimensional random seaway. The model allows the user to prescribe antenna properties (length, diameter, density, etc.), sea conditions (significant wave height, development of sea), tow angle, and tow speed. The model then simulates the antenna-sea interaction for the desired duration to collect statistics relating to antenna performance. The model provides design engineers with a tool to predict antenna performance trends, and to conduct design tradeoff studies. The floating antenna envisioned is for use by a submarine operating at modest speed and depth. / by Sam R. Geiger. / S.M.

CAPES / Nesta dissertação, o uso de antenas direcionais em sistemas de verificação de localização é analisado, considerando um cenário realista para redes veiculares. O objetivo é propor um método para a utilização das antenas direcionais nestes sistemas e verificar quais os benefícios proporcionados em termos de desempenho, que representa a capacidade em diferenciar usuários legítimos e maliciosos. Um sistema que utiliza o modelo de propagação log-normal é inicialmente estudado, seguido de uma analise das alterações necessárias para a utilização de um modelo mais realista. Utilizando este modelo, é demonstrado então como o uso de antenas direcionais pode ser introduzido no sistema, adicionando ao processo de verificação uma etapa adicional. Resultados numéricos são obtidos para verificar as expressões analíticas, seguido das comparações entre os desempenhos de cada um dos sistemas apresentados. Através destes resultados, é constatado que a utilização das antenas direcionais proporciona um aumento considerável no desempenho do sistema e, mesmo quando considerado apenas a etapa adicional de verificação, o desempenho é satisfatório. É feita também uma análise da influência do desvanecimento em pequena escala no sistema, demonstrando qual a relação entre o número de amostras e a porcentagem de erro no desempenho do esquema apresentado. / In this master thesis, a performance analysis of an information-theoretic location verification system using directional antennas is performed, considering a vehicular network scenario. Our goal is to develop a method using directional antennas to improve the system performance. First, a recently introduced scheme that uses a simple propagation model is studied, then it is described the modifications needed when a realistic propagation model is considered. Based on the described scheme, it is shown how the use of directional antennas can be introduced into the system, creating a new validation stage called directional verification. Numerical results are used to validate the analytical expressions, showing also the performance differences between the described systems. Using the obtained results, it is shown that the use of directional antennas increase the system performance and that, even when using only the directional verification stage, the performance is satisfactory. An analysis of the relation between the number of samples and the system's error percentage is also performed, when considered a small scale fading model.

CAPES / Nesta dissertação, o uso de antenas direcionais em sistemas de verificação de localização é analisado, considerando um cenário realista para redes veiculares. O objetivo é propor um método para a utilização das antenas direcionais nestes sistemas e verificar quais os benefícios proporcionados em termos de desempenho, que representa a capacidade em diferenciar usuários legítimos e maliciosos. Um sistema que utiliza o modelo de propagação log-normal é inicialmente estudado, seguido de uma analise das alterações necessárias para a utilização de um modelo mais realista. Utilizando este modelo, é demonstrado então como o uso de antenas direcionais pode ser introduzido no sistema, adicionando ao processo de verificação uma etapa adicional. Resultados numéricos são obtidos para verificar as expressões analíticas, seguido das comparações entre os desempenhos de cada um dos sistemas apresentados. Através destes resultados, é constatado que a utilização das antenas direcionais proporciona um aumento considerável no desempenho do sistema e, mesmo quando considerado apenas a etapa adicional de verificação, o desempenho é satisfatório. É feita também uma análise da influência do desvanecimento em pequena escala no sistema, demonstrando qual a relação entre o número de amostras e a porcentagem de erro no desempenho do esquema apresentado. / In this master thesis, a performance analysis of an information-theoretic location verification system using directional antennas is performed, considering a vehicular network scenario. Our goal is to develop a method using directional antennas to improve the system performance. First, a recently introduced scheme that uses a simple propagation model is studied, then it is described the modifications needed when a realistic propagation model is considered. Based on the described scheme, it is shown how the use of directional antennas can be introduced into the system, creating a new validation stage called directional verification. Numerical results are used to validate the analytical expressions, showing also the performance differences between the described systems. Using the obtained results, it is shown that the use of directional antennas increase the system performance and that, even when using only the directional verification stage, the performance is satisfactory. An analysis of the relation between the number of samples and the system's error percentage is also performed, when considered a small scale fading model.

A comunicação em redes veiculares ad hoc (VANETs) é comumente dividida em dois cenários, chamados veículo-a-veículo (V2V) e veículo-a-infraestrutura (V2I). Objetivando estabelecer uma comunicação segura contra espiões, trabalhos recentes tem proposto a troca de chaves secretas baseado na variação da força do sinal recebido (RSS). Entretanto, o bom desempenho de tal método depende da taxa de variação do canal, sendo mais apropriado a cenários em que o canal varia rapidamente, como geralmente é o caso da comunicação V2V. Já na comunicação V2I, o canal normalmente possui desvanecimento lento. Neste trabalho, é proposta a utilização de múltiplas antenas com o intuito de gerar artificialmente um canal de desvanecimento rápido, permitindo assim a extração de chaves secretas através da RSS em um cenário V2I. Análises numéricas mostram que o modelo proposto pode obter desempenho superior, em termos de taxa de extração de bits secretos, do que o modelo de salto em frequência proposto na literatura. / The communication in vehicular ad hoc networks (VANETs) is commonly divided in two scenarios, namely vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I). Aiming at establishing secure communication against eavesdroppers, recent works have proposed the exchange of secret keys based on the variation in received signal strength (RSS). However, the performance of such scheme depends on the channel variation rate, being more appropriate for scenarios where the channel varies rapidly, as is usually the case with V2V communication. In the communication V2I, the channel commonly undergoes slow fading. In this work we propose the use of multiple antennas in order to artificially generate a fast fading channel so that the extraction of secret keys out of the RSS becomes feasible in a V2I scenario. Numerical analysis shows that the proposed model can outperform, in terms of secret bit extraction rate, a frequency hopping-based method proposed in the literature.

A comunicação em redes veiculares ad hoc (VANETs) é comumente dividida em dois cenários, chamados veículo-a-veículo (V2V) e veículo-a-infraestrutura (V2I). Objetivando estabelecer uma comunicação segura contra espiões, trabalhos recentes tem proposto a troca de chaves secretas baseado na variação da força do sinal recebido (RSS). Entretanto, o bom desempenho de tal método depende da taxa de variação do canal, sendo mais apropriado a cenários em que o canal varia rapidamente, como geralmente é o caso da comunicação V2V. Já na comunicação V2I, o canal normalmente possui desvanecimento lento. Neste trabalho, é proposta a utilização de múltiplas antenas com o intuito de gerar artificialmente um canal de desvanecimento rápido, permitindo assim a extração de chaves secretas através da RSS em um cenário V2I. Análises numéricas mostram que o modelo proposto pode obter desempenho superior, em termos de taxa de extração de bits secretos, do que o modelo de salto em frequência proposto na literatura. / The communication in vehicular ad hoc networks (VANETs) is commonly divided in two scenarios, namely vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I). Aiming at establishing secure communication against eavesdroppers, recent works have proposed the exchange of secret keys based on the variation in received signal strength (RSS). However, the performance of such scheme depends on the channel variation rate, being more appropriate for scenarios where the channel varies rapidly, as is usually the case with V2V communication. In the communication V2I, the channel commonly undergoes slow fading. In this work we propose the use of multiple antennas in order to artificially generate a fast fading channel so that the extraction of secret keys out of the RSS becomes feasible in a V2I scenario. Numerical analysis shows that the proposed model can outperform, in terms of secret bit extraction rate, a frequency hopping-based method proposed in the literature.