Target tracking using Wireless Sensor Networks (WSNs) has drawn lots of attentions after the recent advances of wireless technologies. Target tracking aims at locating one or several mobile objects and depicting their trajectories over time. The applications of Object Tracking Sensor Networks (OSTNs) include but not limited to environmental and wildlife monitoring, industrial sensing, intrusion detection, access control, traffic monitoring, patient monitoring in the health-related studies and location awareness in the battle eld. One of the most rewarding applications of target tracking is wildlife monitoring. Wildlife monitoring is used to protect the animals which are endangered to extinction. Road safety applications are another popular usage of wildlife monitoring using WSNs. In this thesis, the issues and challenges of energy-efficient wildlife monitoring and target tracking using WSNs are discussed. This study provides a survey of the proposed tracking algorithms and analyzes the advantages and disadvantages of these algorithms. Some of the tracking algorithms are proposed to increase the energy e ciency of the tracking algorithm and to prolong the network lifetime; while, other algorithms aim at improving the localization accuracy or decreasing the missing rate. Since improving the energy efficiency of the system provides more alive sensors over time to locate the target; it helps to decrease the missing rate as the network ages. Thus, this study proposes to adjust the sensing radius of the sensor nodes in real-time to decrease the sensing energy consumption and prolong the network lifetime. The proposed VAriable Radius Sensor Activation (VARSA) mechanism for target tracking using wireless sensor networks tackles the energy consumption issues due to resource constraints of the WSNs. VARSA reduces the radio covered area of each sensor node to only cover the Area of Interest (AoI) which is the location of the target in tracking applications. Thus, VARSA aims at decreasing the sensing energy consumption which leads to encreasing the network life time. In addition, VARSA decreases the missing rate over time as it provides more alive sensors to detect the target compared to previous activation algorithms as the network ages. VARSA is compared to PRediction-based Activation (PRA) and Periodic PRediction-based Activation (PPRA) algorithms which are two of the most promising algorithms proposed for sensor activation. The simulation results show that VARSA outperforms PRA and PPRA. VARSA prolongs the lifetime of the network and decreases the missing rate of the target over time.

The recent development of Integrated Power Systems (IPS) for shipboard application has opened the horizon to introduce new technologies that address the increasing power demand along with the associated performance specifications. Similarly, the Shipboard Power System (SPS) features system components with multiple dynamic characteristics and require stringent regulations, leveraging a challenge for an efficient system level management. The shipboard power management needs to support the survivability, reliability, autonomy, and economy as the key features for design consideration. To address these multiple issues for an increasing system load and to embrace future technologies, an autonomic power management framework is required to maintain the system level objectives. To address the lack of the efficient management scheme, a generic model-based holistic power management framework is developed for naval SPS applications. The relationship between the system parameters are introduced in the form of models to be used by the model-based predictive controller for achieving the various power management goals. An intelligent diagnostic support system is developed to support the decision making capabilities of the main framework. Naïve Bayes’ theorem is used to classify the status of SPS to help dispatch the appropriate controls. A voltage control module is developed and implemented on a real-time test bed to verify the computation time. Variants of the limited look-ahead controls (LLC) are used throughout the dissertation to support the management framework design. Additionally, the ARIMA prediction is embedded in the approach to forecast the environmental variables in the system design. The developed generic framework binds the multiple functionalities in the form of overall system modules. Finally, the dissertation develops the distributed controller using the Interaction Balance Principle to solve the interconnected subsystem optimization problem. The LLC approach is used at the local level, and the conjugate gradient method coordinates all the lower level controllers to achieve the overall optimal solution. This novel approach provides better computing performance, more flexibility in design, and improved fault handling. The case-study demonstrates the applicability of the method and compares with the centralized approach. In addition, several measures to characterize the performance of the distributed controls approach are studied.

No description available.

Multi-core and many-core systems are the norm in contemporary processor technology and are expected to remain so for the foreseeable future. Parallel programming is, thus, here to stay and programmers have to endorse it if they are to exploit such systems for their applications. Programs using parallel programming primitives like PThreads or OpenMP often exploit coarse-grain parallelism, because it offers a good trade-off between programming effort versus performance gain. Some parallel applications show limited or no scaling beyond a number of cores. Given the abundant number of cores expected in future many-cores, several cores would remain idle in such cases while execution performance stagnates. This thesis proposes using cores that do not contribute to performance improvement for running implicit fine-grain speculative threads. In particular, we present a many-core architecture and protocols that allow applications with coarse-grain explicit parallelism to further exploit implicit speculative parallelism within each thread. We show that complementing parallel programs with implicit speculative mechanisms offers significant performance improvements for a large and diverse set of parallel benchmarks. Implicit speculative parallelism frees the programmer from the additional effort to explicitly partition the work into finer and properly synchronized tasks. Our results show that, for a many-core comprising 128 cores supporting implicit speculative parallelism in clusters of 2 or 4 cores, performance improves on top of the highest scalability point by 44% on average for the 4-core cluster and by 31% on average for the 2-core cluster. We also show that this approach often leads to better performance and energy efficiency compared to existing alternatives such as Core Fusion and Turbo Boost. Moreover, we present a dynamic mechanism to choose the number of explicit and implicit threads, which performs within 6% of the static oracle selection of threads. To improve energy efficiency processors allow for Dynamic Voltage and Frequency Scaling (DVFS), which enables changing their performance and power consumption on-the-fly. We evaluate the amenability of the proposed explicit plus implicit threads scheme to traditional power management techniques for multithreaded applications and identify room for improvement. We thus augment prior schemes and introduce a novel multithreaded power management scheme that accounts for implicit threads and aims to minimize the Energy Delay2 product (ED2). Our scheme comprises two components: a “local” component that tries to adapt to the different program phases on a per explicit thread basis, taking into account implicit thread behavior, and a “global” component that augments the local components with information regarding inter-thread synchronization. Experimental results show a reduction of ED2 of 8% compared to having no power management, with an average reduction in power of 15% that comes at a minimal loss of performance of less than 3% on average.

Esta dissertação de mestrado propõe uma nova política de distibuição de requisições em cluster de web services, denominada Política Performance. Essa política provê uma distribuição transparente, flexível e dinâmica das requisições na plataforma em que é executada. Um estudo sistemático também é realizado para analisar a qualidade dos índices de carga empregados no contexto de web services e propõe um novo índice capaz de representar fielmente o desempenho dos web services e encapsular a complexidade estrutural da plataforma. Também é proposto em Módulo Gerenciador de Energia capaz de prover sustentabilidade à plataforma, reduzindo o consumo de energia elétrica sem prejudicar a alta confiabilidade na distribuição das requisições e com baixo impacto no tempo médio de resposta. Os estudos experimentais realizados neste trabalho mostraram que a Política Performance permitiu um melhor desempenho no atendimento das requisições realizadas à plataforma. EStes resultados referem-se a um desempenho superior a 70% no tempo médio de resposta, quando comparado ao desempenho demosntrado pela política padrão do Mod_cluster. O Módulo Gerenciador de Energia proporcionou uma redução de aproximadamente 30% no consumo de energia da plataforma mantendo a alta confiabilidade na distribuição das requisições / The Master\'s dissertation proposes a new for distribution of requests in cluster of web services, named policy Performance. This policy provides a transparent flexible and dynamic distribution of requests on the plataform. A systematic study is also conducted to example the quality of load indices used in the context of web services., and proposes a new index that accurately represent the performance of web services and encapsulate the complexity structural of the plataform. Also proposed is an Energy Manager Module capable of providing sustainability to the plataform, reducing power consumption without sacrificing high reliability in the distribution of request and low impact on the average response time. Our main results show that policy Performance has a better performance in handling requests sent to the plataform. Our results show a gain of performance higher than 70% in average responswetime when compared to the perormance demonstrated by the defaut policy Mod_cluster. The Power Manager Module reduced byapproximately 30% the energy consumption of the plataform even keeping the high reliability in the distribution of requests

In recent years, due to the combination of technological advances in the fields ofmeasurement instrumentation, communications, home-health care and textile-technology thedevelopment of medical devices has shifted towards applications of personal healthcare.There are well known the available solutions for heart rate monitoring successfully providedby Polar and Numetrex. Furthermore new monitoring applications are also investigated. Amongthese non-invasive monitoring applications, it is possible to find several ones enable bymeasurements of Electrical Bioimpedance.Analog Devices has developed the AD5933 Impedance Network Analyzer which facilitatesto a large extent the design and implementation of Electrical Bioimpedance Spectrometers in amuch reduced space. Such small size allows the development of a fully wearable bioimpedancemeasurement.With the development of a Electrical Bioimpedance-enable wearable medical device in focusfor personal healthcare monitoring, in this project, the issue of power management has beentargeted and a battery-driven Electrical Bioimpedance Spectrometer based in the AD5933 hasbeen implemented. The resulting system has the possibility to operate with a Li-Po battery with apower autonomy over 17 hours.

Une grande partie des nouvelles générations d'objets connectés ne pourra se développer que s'il est possible de les rendre entièrement autonomes sur le plan énergétique. Même si l'utilisation de batteries ou de piles résout une partie de ce problème en assurant une autonomie qui peut-être importante avec des coûts relativement faibles, elle introduit non seulement des contraintes de maintenance incompatibles avec certaines applications, mais aussi des problèmes de pollution. La récupération de l'énergie thermique, mécanique, électromagnétique, solaire ou éolienne est une solution très prometteuse. Dans ce cas, la vie de l'objet connecté peut-être prolongée. Cependant, l'énergie récupérée dépend fortement des conditions au voisinage du dispositif et peut donc varier de façon périodique ou aléatoire. Il parait alors important d'adapter le système (mesure et transmission de données) aux contraintes de la récupération d'énergie. L'objectif de la thèse est de proposer une solution de capteur autonome basée sur un système de récupération et de gestion multisources d'énergies (solaire et éolienne) et pouvant-être mis en oeuvre dans différentes classes d'applications IoT. On s'intéresse, dans un premier temps, à la modélisation de la consommation du noeud capteur. Ensuite, on modélise le système de récupération multisources. Puis, on se focalise sur le management de puissance du système autonome. Ce management est basé sur des prédictions de l'énergie disponible et de celle consommée. Enfin, le travail de modélisation et d'optimisation est validé par des expérimentations afin d’avoir un démonstrateur de Capteur Communicant Autonome en Énergie pour les applications IoT. / Researchers aim to develop entirely autonomous sensors. By ensuring an important autonomy, the use of batteries solves part of the energy problem with relatively low costs. However, batteries introduce different problems such as maintenance and environmental pollution. Harvesting thermal, mechanical, electromagnetic, solar or wind energy present in the environment is an attractive solution. Using harvested energy from their surroundings, wireless sensor nodes can significantly increase their typical lifetime. Nevertheless, the harvested energy depends on the surrounding conditions of the device and can vary periodically or randomly. It seems important to adapt the system (measurement and data transmission) to the harvesting energy constraints. The thesis objective is to provide an autonomous sensor solution based on a multisources energy harvesting and management system (solar and wind energies), which can be used in different IoT applications. First, we are interested in modeling and optimizing the sensor node energy consumption. Then, the multiple harvesting system is modeled according to the energy needs of the sensor node. Besides, we focus on the power management of the autonomous system. This management part is based on predictions of both available and consumed energies. Finally, the proposed modeling and optimization studies are validated with experimental works in order to develop an Autonomous Communicating Sensor platform for IoT applications.

Smart grid has been described as the Energy Internet: Where Energy Technology meets Information Technology. The incorporation of such technology into vast existing utility infrastructures offers many advantages, including possibilities for new smart appliances, energy management systems, better integration of renewable energy, value added services, and new business models, both for supply- and demand-side management. Smart grid also replaces aging utility technologies that are becoming increasingly unreliable, as the average ages for many critical components in utility systems now exceed their original design lives. However, while smart grid offers the promise of revolutionizing utility delivery systems, many questions remain about how such systems can be rolled out at the state, regional, and national levels. Many unique regulatory and market structure challenges exist, which makes it critical to pick the right technology for the right situation and to employ it in the right manner. Technology Roadmapping may be a valuable approach for helping to understand factors that could affect smart grid technology and product development, as well as key business, policy and regulatory drivers. As emerging smart grid technologies are developed and the fledgling industry matures, a critical issue will be understanding how the combination of industry drivers impact one another, what barriers exist to achieving the benefits of smart grid technologies, and how to prioritize R&D and acquisition efforts. Since the planning of power grids often relies on regional factors, it will also be important investigate linkages between smart grid deployment and regional planning goals. This can be used to develop strategies for overcoming barriers and achieving the benefits of this promising new technology. This research builds upon existing roadmapping processes by considering an integrated set of factors, including policy issues, which are specifically tuned to the needs of smart grids and have not generally been considered in other types of roadmapping efforts. It will also incorporate expert judgment quantification to prioritize factors, show the pathways for overcoming barriers and achieving benefits, and discussing the most promising strategies for achieving these goals.

Mobile sensing applications (MSAs) are an emerging class of applications that process continuous sensor data streams to make time-sensitive inferences. Representative application domains range from environmental monitoring, context-aware services to recognition of physical activities and social interactions. Example applications involve city air quality assessment, indoor localization, pedometer and speaker identification. The common application workflow is to read data streams from the sensors (e.g, accelerometers, microphone, GPS), extract statistical features, and then present the inferred high-level events to the user. MSAs in the healthcare domain especially draw a significant amount of attention in recent years because sensor-based data collection and assessment offer finer-granularity, timeliness, and higher accuracy in greater quantity than traditional, labor-intensive, data gathering mechanisms in use today, e.g., surveys methods. The higher fidelity and accuracy of the collected data expose new research opportunities, improve the reliability and accuracy of medical decisions, and empower users to manage personal health more effectively. Nonetheless, a critical challenge to practical deployment of MSAs in real-world is to effectively manage limited resources of mobile platforms to meet stringent quality of service (QoS) requirements in terms of processing throughput and delay while ensuring long term robustness. To address the challenge, we model MSAs in dataflows as a graph of processing elements that are connected by communication channels. The processing elements may execute in parallel as long as they have sufficient data to process. A key feature of the dataflow model is that it explicitly capture parallelism and data dependencies between processing elements. Based on the graph composition, we first proposed CSense, a stream-processing toolkit for robust and high-rate MSAs. In this work, CSense provide a simple language for developers to describe their sensing flow without the need to deal with system intricacy, such as memory allocation, concurrency control and power management. The results show up to 19X performance difference may be achieved automatically compared with a baseline using the default runtime concurrency and memory management. Following this direction, we saw the opportunities that MSAs can be significantly improved from the perspective of memory performance and energy efficiency in view of the iterative execution. Therefore, we next focus on optimizing the runtime memory management through compile time analysis. The contribution is a stream compiler that captures the whole program memory behavior to generate an efficient memory layout for runtime access. Experiments show that our memory optimizations reduce memory footprint by as much as 96% while matching or improving the performance of the StreamIt compiler with cache optimizations enabled. On the other hand, while there is a significant body of work that has focused on optimizing the throughput or latency of processing sensor streams, little to no attention has been given to energy efficiency. We proposed an accurate offline energy prediction model for MSAs that leverages the pipeline structure and iterative execution nature to search for the most energy saving batching configuration w.r.t. a deadline constraint. The developers are expected to visualize the energy delay trade-off in the parameter space without runtime profiling. The evaluation shows the worst-case prediction errors are about 7% and 15% for energy and latency respectively despite variable application workloads.

Design of computing systems — from pocket-sized smart phones to massive cloud based data-centers — have one common daunting challenge : minimizing the power consumption. In this effort, power management sector is undergoing a rapid and profound transformation to promote clean and energy proportional computing. At the hardware end of system design, there is proliferation of specialized, feature rich and complex power management hardware components. Similarly, in the software design layer complex power management suites are growing rapidly. Concurrent to this development, there has been an upsurge in the integration of third-party components to counter the pressures of shorter time-to-market. These trends collectively raise serious concerns about trust and security of power management solutions. In recent times, problems such as overheating, performance degradation and poor battery life, have dogged the mobile devices market, including the infamous recall of Samsung Note 7. Power outage in the data-center of a major airline left innumerable passengers stranded, with thousands of canceled flights costing over 100 million dollars. This research examines whether such events of unintentional reliability failure, can be replicated using targeted attacks by exploiting the security loopholes in the complex power management infrastructure of a computing system. At its core, this research answers an imminent research question: How can system designers ensure secure and reliable operation of third-party power management units? Specifically, this work investigates possible attack vectors, and novel non-invasive detection and defense mechanisms to safeguard system against malicious power attacks. By a joint exploration of the threat model and techniques to seamlessly detect and protect against power attacks, this project can have a lasting impact, by enabling the design of secure and cost-effective next generation hardware platforms.