【作者】 祝启龙；

【导师】 林蔚；

【作者基本信息】 哈尔滨工程大学 ， 应用数学， 2011， 硕士

【摘要】 无线传感器网络是由许多无线传感器节点以多跳路由、无线通信的方式组成的网络。无线传感器负责采集信息,向基站传送数据。在基站对数据进行分析、处理、决策。由于无线传感器网络的成本低、精度高、多跳路由,可实行远程监控等优点,因此在许多方面得到应用。除外,它的自组织分布式的特点使得它在众多领域有着广泛的发展前景。无线传感器节点主要由电池供电。由于每个节点具有较少的能量,较弱通信能力和计算能力,部署的面积广,所以在不更换能源的前提下增加网络生存时间就变得较为复杂。节约网络能量成为网络研究的一个方向。本文首先叙述了课题研究的意义、课题来源和国内外研究状况及本文的主要工作。叙述了无线传感器网络的基本知识和目前在数据处理方面常用的处理办法。根据该网络的特殊情况对数据融合问题展开讨论,比较了现有数据融合方法对处理该网络节能问题的优点和存在的问题。其次,根据无线传感器网络测量数据在时间和空间上的相似性,本论文提出数据融合算法-—TFA算法。该算法通过对传感器节点设定阈值来控制传感器节点的数据传输量,进而在传感器节点处,对采集的数据进行融合；再在控制中心处,对采集的数据进行补偿和拟合。TFA算法在解决了节能问题的同时提高了数据融合的精度。仿真结果表明,TFA算法与主成分分析法、可靠性分析法和模糊加权等一步的算法相比更节省能量,数据精度更高。更多还原

【Abstract】 Wireless sensor network is a net composed of many wireless sensor nodes in multi-hop routing, wireless communication way. Wireless sensor is responsible for collecting information, transmit data to the base station, and analysis, process, and make decision on the data in the base station. Since wireless sensor networks, low cost, high precision, multi-hop routing, implementation of remote monitoring, etc., so it can be applied in many ways. Besides, the feature of self-organization distributed makes it broad prospects for development in many fields.Wireless sensor nodes are mainly powered by the battery. Each node has less energy, less communication and numeric, wide area deployment. This makes more complex to increase the survival time of network without the replacement of energy. Then how to save the network energy becomes a topic in the research of network.This paper describes the significance of the research, issues sources and the main work status of domestic and foreign. We describe the basics of wireless sensor networks and processing approach currently used in data. According to the special circumstances of the network, we discuss issues of data fusion, compare the merits and problems in dealing with energy-saving of network by the existing data fusion.Secondly, based the similarity of measurement in time and space on wireless sensor network data, we propose a data fusion algorithm-TFA algorithm. The algorithm controls the amount of data transmission of sensor nodes by setting thresholds on sensor nodes, then makes fusion on the collected data at the sensor nodes, and finally compensates and fit the collected data in the control center.While improving the accuracy of data fusion, TFA algorithm solves the energy problem. Simulation results show that TFA algorithm saves more energy than principal component analysis, reliability analysis and fuzzy weighted algorithm, and has higher data precision.更多还原