【作者】 于平；

【导师】 杨永健；

【作者基本信息】 吉林大学 ， 网络与信息安全， 2014， 硕士

【摘要】 无线传感器网络（Wireless Sensor Network）是一种由大量的传感器节点组成的网络结构。它的构建通常是依赖于传感器节点之间的自组织特性。传感器节点之间通过无线收发模块相互感知、通信和协作，来完成数据采集、分析等工作。传感器节点通常具有成本低、体积小、功耗低等特点，并且随着工业制造水平的提升，传感器节点正向着成本更低、体积更小的方向发展。体积小是传感器节点的优势，但同时也不可避免地带来了其许多非功能性缺陷，通常传感器节点的计算能力较低，存储能力不高，尤其是其能量十分有限。而无线传感器网络最重要的性能标准就是网络生存时间。这就要求在应用中使用无线传感器网络时，必须提出一定的方法来延长网络生命周期。拓扑控制问题的研究也主要针对于这种背景展开。已经有许多的学者在WSNs拓扑控制方面有过大量的研究。但综合观察这些算法，大部分是针对于静态无线传感器网络环境。也就是说，这些研究关注的是节点静止，网络拓扑不变的无线传感器网络。而现实中却存在着许多可以抽象为移动传感器网络的应用，如车载网络等。由此可见，提出一种适用于动态传感网络的拓扑控制算法意义重大。本文的主要工作有以下几点：首先，本文对无线传感器网络的背景知识进行了一定的介绍，包括，无线传感器网络的体系结构、网络特性和关键技术等，重点介绍了静态传感网络的经典拓扑控制算法——LEACH算法。介绍了移动传感器网络方面的背景知识。其次，针对于现有的以LEACH算法为代表的应用于静态网络结构的拓扑控制算法所体现出的不足，提出了一种适用于移动传感器网络的自适应网格化分簇拓扑控制算法AREAL。该算法改进了静态LEACH算法，将节点剩余能量加入到选择簇头的参考因素，从而避免剩余能量过低的节点成为簇头，承担繁重的融合和转发等消耗能量的工作；将移动无线传感器网络区域网格化，分而治之，避免了构建出来的簇大小不一，导致结构失去平衡，不利于负载均衡，算法可以将静态节点推广到动态节点。综上所述，在选择簇头阶段，我们将综合考虑节点剩余能量和节点的瞬时位置，而非像LEACH算法一样的随机选择。采用上述多种方式，我们可以将AREAL很好地应用到了移动网络中。最后，通过仿真实验的方式验证了AREAL算法的性能。将AREAL的实验结果与LEACH和RCM算法相对比。比较了三者之间吞吐量、网络生存时间、能量负载等网络性能指标并给出实验分析。诚然，动态传感器网络拓扑控制方面的研究还属于一个比较新的研究领域。本文也只是针对其在目前的WSNs领域研究较少的背景，尝试提出一种方式来适合动态WSNs的拓扑控制。因此，本算法还有许多不足和可以改进的地方。希望更多的学者能够注意到这一领域的发展前景。更多还原

【Abstract】 WSNs (Wireless Sensor Network) is a network structure composed of a large number ofsensor nodes. It is usually dependent on the construction of self-organization among sensornodes. The sensor nodes recognize, communicate and collaborate with each other through thewireless transceiver module, to complete the data collection, analysis and so on.Sensor nodes are usually low-cost, small and with low-power consumption. With theupgrading of the industrial manufacturing level, the sensor nodes are becoming lower-cost,smaller. Small size is an advantage of the sensor nodes, but also inevitably brings many of itsnon-functional defects. Generally sensor nodes have lower computing ability, and the storagecapacity is not high, especially its energy is very limited. The most important performance ofwireless sensor network is criteria network lifetime. This requires that we must take certainways to extend the network life cycle by the use of wireless sensor networks in an application.Study of the topology control is also to aim at the objective.There have been many scholars in the field of WSNs topology control making lots ofresearch. But comprehensive observation of these algorithms, mostly for static wireless sensornetwork environment. In other words, these studies concern is wireless sensor networks withstationary nodes. But in reality, there are many possible abstract for mobile sensor networkapplications, such as in-vehicle networks. Thus, the study on topology control of a dynamicsensor network is significant.The main work of this paper are the following:Firstly, it introduces the background knowledge of wireless sensor networks, includingthe architecture, network characteristics and key technologies such as wireless sensornetworks, focusing on the classic topology control algorithm static sensor networks LEACHalgorithm. Describes the background knowledge of mobile sensor networks.Secondly, the existing topology control algorithm represented LEACH algorithm appliedto a static network structure embodied deficiencies, we propose a suitable mobile sensornetworks, adaptive grid clustering topology control algorithm AREAL. The algorithmimproves the static LEACH algorithm, the node is added to select the remaining energy of cluster head reference factor, thus avoiding the residual energy of cluster head node becomestoo low, the work bears the heavy consumption of energy integration and forwarding; mobilewireless sensor networks regional grid, divide and conquer, to avoid the building up ofclusters of different sizes, resulting in loss of balance structure is not conducive to the loadbalancing algorithm can be static nodes to promote the dynamic node. In summary, thechoice of cluster head phase, we will consider the residual energy and the instantaneousposition of the nodes, not the same as LEACH algorithm randomly selected. With the abovein many ways, we can AREAL well applied to the mobile network.Finally, we verify the performance of AREAL algorithm by way of simulations. Theexperiment results AREAL RCM algorithm with LEACH and the relative ratio. Compared thethroughput, network lifetime, energy load and other network performance between the threeexperimental analysis is given.Indeed, the study of dynamic sensor network topology control also belongs to arelatively new field of study. This article is only for its research in the field in the currentWSNs background; try to propose a way to fit a dynamic topology control WSNs. Therefore,the algorithm has many shortcomings and areas for improvement. Hope that more scholarsable to note the development prospects in this field.更多还原