基于部分信标失效的水下无线传感器网络定位技术的研究

发布时间：2018-06-29 00:42

本文选题：水下无线传感器网络 + 错误信标筛选　；参考：《南京邮电大学》2017年硕士论文

【摘要】：水下无线传感器网络由部署在监测水域的低成本、自组织的智能传感器组成以去协同完成水域环境监测的任务,在环境监测、无人监控深海区域、自然灾难预防、军事预防等领域有着广泛的应用。而在大部分实际应用中,采集的数据必须标记位置信息时才构成具体物理意义。此外,准确的节点位置信息有助于提高路由协议的效率,优化网络的拓扑设计及均衡网络不同区域的能耗,因此对水下无线传感器网络中的定位算法进行研究是非常必要的。在复杂而恶劣的水下环境中,由于水流的变化、水下生物触碰、强烈的电磁波干扰等原因,部分信标节点很容易移动或损坏,导致这些信标失效或其位置信息可能出错,而在这些错误(或失效)信标的辅助下,普通节点的定位误差显然会增大。针对错误(或失效)信标问题,本文分别提出了两种基于错误信标筛选的定位算法。本文首先提出了一种基于k均值的错误信标筛选算法,在周围信标节点的帮助下,每一个信标节点使用改进的三边测量法去定位,然后利用k均值聚类算法筛选出具有最大定位误差的信标节点,剩余的信标节点继续迭代以上过程,直到每个信标节点的估计误差不超过预设阀值。另外,针对k均值聚类算法的分类准确性很大程度依赖初始中心值并且分类结果易陷入局部最优的问题,本文将粒子群算法和k均值算法相结合,又提出了一种水下无线传感器网络中基于粒子群聚类的错误信标筛选算法,可以更加准确地筛选出错误信标。最后本文对上述算法进行了复杂度分析和仿真测试。结果表明本文所提出的算法能够有效地筛选出错误信标,从而为普通节点提供一个可靠的信标集合,最终达到提高普通节点定位精度的目的。
[Abstract]:Underwater wireless sensor networks are made up of low-cost, self-organizing intelligent sensors deployed in monitoring waters to work together to perform environmental monitoring tasks in waters, in environmental monitoring, in unmonitored deep-sea areas, and in natural disaster prevention. Military prevention and other fields have a wide range of applications. In most practical applications, the collected data must mark the location information to form a specific physical meaning. In addition, accurate node location information is helpful to improve the efficiency of routing protocol, optimize the topology design of the network and equalize the energy consumption in different areas of the network. Therefore, it is very necessary to study the localization algorithm in underwater wireless sensor networks. In the complex and bad underwater environment, some beacon nodes are easily moved or damaged due to the change of water flow, underwater biological contact, strong electromagnetic interference and so on, resulting in the failure of these beacons or the possible errors in their location information. With the aid of these false (or invalid) beacons, the location errors of common nodes will obviously increase. Aiming at the problem of error (or failure) beacon, two localization algorithms based on error beacon filtering are proposed in this paper. In this paper, an error beacon selection algorithm based on k-means is proposed. With the help of the surrounding beacon nodes, each beacon node uses an improved trilateral measurement method to locate the beacon. Then the k-means clustering algorithm is used to select the beacon nodes with the maximum location error, and the remaining beacon nodes continue to iterate the above process until the estimation error of each beacon node does not exceed the preset threshold. In addition, aiming at the problem that the classification accuracy of k-means clustering algorithm depends on the initial center value to a great extent and the result of classification is prone to fall into local optimum, the particle swarm optimization algorithm and the k-means algorithm are combined in this paper. An error beacon selection algorithm based on particle clustering in underwater wireless sensor networks is proposed, which can screen error beacons more accurately. Finally, the complexity of the algorithm is analyzed and simulated. The results show that the algorithm proposed in this paper can effectively screen out error beacons, thus providing a reliable set of beacons for ordinary nodes, and finally achieving the purpose of improving the positioning accuracy of common nodes.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.3

【参考文献】