外文翻译原文模板.doc

1、外文资料翻译内容要求：外文资料的内容应为本学科研究领域，并与毕业设计（论文）选题相关的技术资料或专业文献，译文字数应不少于 3000 汉字以上，同时应在译文末注明原文的出处不可采用网络中直接有外文和原文的2、外文资料翻译格式要求：译文题目采用小二号黑体，居中；译文正文采用宋体小四号，段前、段后距为 0 行；行距：固定值 20 磅英文原文如果为打印的话用新罗马（Times New Roman）小四号字装订时原文在前，译文在后文章中有引用的地方在原文中也要体现参考文献也要翻译成中文！An Energy-Efficient Cooperative Algorithm for Data Estimation in Wireless Sensor NetworksAbstract – In Wireless Sensor Networks (WSN), nodes operate on batteries and network’s lifetime depends on energy consumption of the nodes. Consider the class of sensor networks where all nodes sense a single phenomenon at different locations and send messages to a Fusion Center (FC) in order to estimate the actual information. In classical systems all data processing tasks are done in the FC and there is no processing or compression before transmission. In the proposed algorithm, network is divided into clusters and data processing is done in two parts. The first part is performed in each cluster at the sensor nodes after local data sharing and the second part will be done at the Fusion Center after receiving all messages from clusters. Local data sharing results in more efficient data transmission in terms of number of bits. We also take advantage of having the same copy of data at all nodes of each cluster and suggest a virtual Multiple-Input Multiple-Output (V-MIMO) architecture for data transmission from clusters to the FC. A Virtual-MIMO network is a set of distributed nodes each having one antenna. By sharing their data among themselves, these nodes turn into a classical MIMO system. In the previously proposed cooperative/virtual MIMO architectures there has not been any data processing or compression in the conference phase. We modify the existing VMIMO algorithms to suit the specific class of sensor networks that is of our concern. We use orthogonal Space-Time Block Codes (STBC) for MIMO part and by simulation show that this algorithm saves considerable energy compared to classical systems.I. INTRODUCTIONA typical Wireless Sensor Network consists of a set of small, low-cost and energy-limited sensor nodes which are deployed in a field in order to observe a phenomenon and transmit it to a Fusion Center (FC). These sensors are deployed close to one another and their readings of the environment are highly correlated. Their objective is to report a descriptive behavior of the environment based on all measurements to the Fusion Center. This diversity in measurement lets the system become more reliable and robust against failure. In general, each node is equipped with a sensing device, a processor and a communication module (which can be either a transmitter or transmitter/receiver). Sensor nodes are equipped with batteries and are supposed to work for a long period of time without battery replacement. Thus, they are limited in energy and one of the most important issues in designing sensor networks will be the energy consumption of the sensor nodes. To deal with this problem, we might either reduce the number of bits to be transmitted by source compression or reduce the required power for transmission by applying advanced transmission techniques while satisfying certain performance requirement.A lot of research has been done in order to take advantage of the correlation among sensors’ data for reducing the number of bits to be transmitted. Some are based on distributed source coding[1] while others use decentralized estimation[2-5]. In [1], authors present an efficient algorithm that applies distributed compression based on Slepian – Wolf[14] encoding technique and use an adaptive signal processing algorithm to track correlation among sensors data. In [2-5] the problem of decentralized estimation in sensor networks has been studied under different constraints. In these algorithms, sensors perform a local quantization on their data considering that their observations are correlated with that of other sensors. They produce a binary message and send it to the FC. FC combines these messages based on the quantization rules used at the sensor nodes and estimates the unknown parameter. Optimal local quantization and final fusion rules are investigated in these works. The distribution of data assumed for sensor observation in these papers has Uniform probability distribution function. In our model we consider Gaussian distribution introduced in [17] for sensor measurements which is more likely to reality.As an alternative approach, some works have been done using energy-efficient communication techniques such as cooperative/virtual Multiple-Input Multiple-Output (MIMO) transmission in sensor networks [6-11]. In these works, as each sensor is equipped with one antenna, nodes are able to form a virtual MIMO system by performing cooperation with others. In [6] the application of MIMO techniques in sensor networks based on Alamouti[15] space-time block codes was introduced. In [8,9] energy-efficiency of MIMO techniques has been explored analytically and in [7] a combination of dis。