面向水环境监测的传感网覆盖算法研究

发布时间：2018-01-23 22:09

本文关键词： 水下传感器网络无线传感网覆盖异构传感网虚拟力算法最优化算法　出处：《南京邮电大学》2015年博士论文　论文类型：学位论文

【摘要】：水环境监测网络由包含有多种水环境监测传感器的节点组成,可应用在海洋探测,水污染监测,沿海开发,灾难预警,导航定位及军事监控等方面。网络覆盖是无线传感器网络的基本问题之一,它反映了网络节点对无线传感器网络部署区域的监控程度,将在很大程度上影响网络的成本及性能。本文以湖泊、河流等水体的环境监测为背景,构建了水下传感器网络实验环境,重点研究水下传感器网络的覆盖问题。论文针对面向水环境监测的传感网特点设计了水面二维覆盖算法,水下三维覆盖算法及节点的定位算法,通过仿真验证算法的有效性和可行性。本文的主要工作如下：(1)定向移动的覆盖优化算法针对水下传感器随机部署时网络覆盖率较低的问题,以提高网络覆盖率为研究目标,设计出适用于定向移动传感网模型的覆盖算法。算法的核心内容为定义节点相对邻居节点和相对区域边界的理想位置及节点到达理想位置的虚拟距离,根据节点间的位置关系,依次计算网络中节点相对所有邻居节点及边界的虚拟距离,并且将节点相对邻居节点和边界的虚拟距离加权求和作为节点实际移动的距离,对全网中节点的位置进行微调。算法多次迭代运行直至网络中节点位置达到稳定。(2)基于采样的水下传感网覆盖算法以水下三维传感器网络为研究对象,以提高水下三维传感网覆盖为目标,结合采样统计学思想和最优化算法,提出一种基于采样的水下三维覆盖优化算法。具体策略为,首先对三维水下无线传感器节点的部署区域进行平面采样,采样平面与节点的感知圆球相交在平面上形成半径不等的感知圆,将三维空间覆盖问题转化为异构网的平面覆盖问题；然后对平面进行直线采样,将平面覆盖优化问题转化为直线段的覆盖优化问题。以节点移动距离最小为优化目标,采样直线的最大覆盖为约束条件,根据直线段与平面感知圆的交点坐标之间关系,建立数学模型,在节点移动距离最小的情况下达到对直线段的最优覆盖。当采样平面中多条采样直线段达到最优覆盖时,采样平面的覆盖可得到优化；三维空间中多个采样平面的覆盖得到优化,水下三维空间的覆盖也将得到有效优化。仿真结果证实了算法可以有效的提高网络的覆盖率。(3)圆形区域内基于极坐标的覆盖优化算法水面节点的位置直接影响水下三维覆盖算法的性能,针对水环境监测网络随机部署时水面节点分布不均匀问题,提出一种基于极坐标的覆盖优化算法。算法分解为径向优化和圆周方向的优化两部分,对二维水平面的网络节点位置进行优化提高网络的覆盖率。具体策略为,径向优化时节点仅可以沿原点到节点的连线移动,计算节点相对每个邻居节点的虚拟半径大小,虚拟半径的矢量和为节点沿半径实际移动的距离；圆周方向优化时,节点仅可沿极坐标系中节点所在的圆环移动,计算节点相对每个邻居节点的虚拟角度的大小,虚拟角度的矢量和为节点沿圆周方向实际移动的角度。按照节点ID顺序依次迭代优化,并通过节点极坐标位置中半径值的限制,简单有效的将节点控制在部署区域内,尽可能减少节点间的重叠区域,优化网络的覆盖率。(4)面向水环境监测的定位算法针对恶劣环境中不适合放置过多的信标节点,或者信标节点失效时,如何部署信标节点对水环境中的传感器节点定位的问题,设计一种基于自适应网格的迭代定位算法对未知节点进行定位。具体策略为,将信标节点放置在未知节点分布区域的边缘,采用三角形余弦定理先对部分区域的未知节点进行定位,在已定位节点中选取最佳的信标节点,根据信标节点的位置和通信半径计算下一网格的宽度或者高度,迭代定位直至未知节点全部被定位。为了提高定位算法的适用性,对算法进行了改进,新的算法结合了极大似然定位算法和三角余弦定理定位算法,将信标节点按照一定的规则放置在节点部署区域的边缘。根据三角形余弦定理估计可定位区域内的所有未知节点的位置,已获得位置信息的节点标记为信标节点。计算剩余未知节点的邻居节点中包含信标节点的数目,若含有3个及以上的信标节点,则执行极大似然定位算法,迭代定位直到所有的节点被定位。
[Abstract]:Water environment monitoring network consists of nodes containing a variety of water environment monitoring sensor, can be used in ocean exploration, water pollution monitoring, coastal development, disaster warning, navigation and military surveillance etc.. Network coverage is one of the fundamental problems in wireless sensor networks, it reflects the degree of regional monitoring network node of wireless sensor network the deployment will affect the cost and performance of the network to a great extent. In this paper, lakes, rivers and other water environmental monitoring as the background, constructs the experimental environment of underwater sensor networks, coverage problem of underwater sensor networks. The thesis focuses on the key features of sensor network for water environment monitoring design of the surface two-dimensional covering algorithm, positioning three dimensional coverage algorithm and node algorithm under water, and validated through simulation and feasibility of the algorithm. The main work of this paper are as follows: (1) directional movement Coverage optimization algorithm for underwater sensor random deployment network coverage is low, in order to improve the network coverage rate as the research object, design a covering algorithm suitable for model oriented mobile sensor network. The core algorithm for the ideal location defined node relative neighbor nodes and relative area boundary and node to reach the ideal position of virtual distance according to the position relationship between the nodes, the nodes in the network are calculated relative to all neighbor nodes and the virtual boundary distance and node relative neighbor nodes and the boundary of the virtual distance weighted sum for the actual node moving distance, the fine-tuning of the nodes in the whole network location. Node location algorithm to achieve stable operation until the network. (2) the sampling of underwater sensor network coverage algorithm to 3D underwater sensor network as the research object based on, in order to improve the underwater 3D transmission The sense of network coverage as the goal, combined with statistical sampling idea and optimization algorithm, proposed a sampling of underwater 3D coverage optimization algorithm based on the specific strategy for the first, the deployment area of 3D underwater wireless sensor node plane sampling, sampling plane and node sensing sphere intersected forming radius in the plane circle ranging from perception and the 3D space coverage problem is transformed into plane heterogeneous network coverage problem; then the plane linear sampling plane coverage optimization problem into the coverage optimization problem of line segments. The minimum node moving distance, the maximum coverage sampling line as the constraint conditions, according to the relationship between the coordinates of the intersection line segments and plane perception round, the establishment of mathematical model, under the condition of minimum optimal on line segment coverage in mobile node distance. When the sampling plane multiple sampling line Reach the optimal coverage, the coverage of sampling plane can be optimized in three-dimensional space; a plurality of sampling plane coverage has been optimized, three-dimensional underwater coverage will be effectively optimized. The simulation results show that the algorithm can effectively improve the network coverage rate. (3) the circular area coverage optimization based on polar coordinates the location of the node algorithm directly affects the performance of 3D coverage algorithm for underwater, aiming at the problem of water environment monitoring network node random deployment of water distribution is not uniform, proposed a coverage optimization algorithm based on polar coordinates. The two part is divided into radial and circumferential optimization optimization direction optimization algorithm, increase network coverage rate of two-dimensional level the network node location. Specific strategies for optimization, radial node to node connection can only move along the origin node relative to each neighbor node virtual half The size of the vector, and virtual radius nodes along the radius of the actual moving distance; the circumferential direction optimization, nodes can only move along the ring node in polar coordinates, calculate each neighbor node node relative virtual angle size, vector angle and virtual node for the actual moving along the circumferential direction angle. According to the ID sequence of nodes and nodes through iterative optimization, polar coordinate position radius value, simple and effective control in the node deployment area, as far as possible to reduce the overlap area between the nodes, optimize the network coverage rate. (4) localization algorithm for water environment monitoring in harsh environments for place the number of beacon nodes, or beacon node failure, how to deploy the beacon nodes to locate sensor nodes in the water environment problem, design an iterative localization algorithm based on adaptive mesh on Locate the unknown node. Specific strategies for the beacon nodes are placed at the edge of the unknown node distribution, using triangle cosine theorem to locate the unknown nodes in some areas, has selected the best location of the beacon node node, calculating the next grid width or height according to the beacon node position and the communication radius. Iterative localization of unknown nodes until all are located. In order to improve the applicability of the algorithm, an improved algorithm, the new algorithm combines the maximum likelihood localization algorithm and triangle cosine theorem localization algorithm, the beacon nodes in accordance with certain rules placed on the edge of the deployment area nodes. According to the triangle cosine theorem can estimate all unknown node location regional location, node markers have been obtained the location information of the beacon node. The neighbor nodes calculate the remaining unknown nodes contained in the letter The number of standard nodes, if it contains 3 or more beacon nodes, performs the maximum likelihood location algorithm and iteratively locate until all nodes are located.

【学位授予单位】：南京邮电大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：X84;TP212

【相似文献】