无线传感器网络移动节点三维定位算法研究

发布时间：2018-10-15 07:37

【摘要】：位置信息对传感器网络的检测活动至关重要。在实际应用中,传感器节点往往分布在三维区域中,如山区、森林、海洋等,且节点往往具有不可忽略的移动性。因此,这就要求定位算法能够实现移动节点在三维空间中的定位。本文针对移动节点在三维空间中的特性,设计了一种基于蒙特卡洛的无线传感器网络移动节点非测距分布式锚箱三维定位算法。算法利用外接正方体来表示节点的通信范围和移动范围,根据未知节点每时隙最大移动范围和锚节点通信范围形成采样区域,依据锚箱体积确定采样数量,随机采集的样点取均值作为未知节点的估计位置。仿真表明,通过不同采样数量、时隙、锚节点比例、节点数量、节点移动速度、通信半径以及通信范围的不规则度对定位误差影响的分析,在一定节点移动速度、较小的通信半径和锚节点比例下算法取得了较理想的定位精度,且优于质心算法,有效实现了三维环境中移动节点的定位。通过引进邻居节点有效样点投票机制,本文还研究了一种加权蒙特卡洛移动节点的三维定位算法。算法充分利用邻居节点的位置信息,给筛选之后的有效样点进行投票,给更接近真实位置的样本分配更大的权值,通过权值来反映邻居节点对估计位置的影响提高定位精度。仿真显示,通过不同时隙、锚节点比例、节点数量、节点移动速度和通信半径对定位误差影响的分析,算法定位精度高于锚箱三维算法和质心算法。该算法具有较高的鲁棒性、可扩展性和定位精度,且是一种无需测距的分布式三维定位算法,可在移动无线传感器网络中大规模的实际应用。
[Abstract]:Location information is very important for sensor network detection. In practical applications, sensor nodes are often distributed in three-dimensional regions, such as mountains, forests, oceans and so on, and the nodes often have the mobility that can not be ignored. Therefore, the localization algorithm is required to realize the location of mobile nodes in three-dimensional space. In view of the characteristics of mobile nodes in 3D space, this paper designs a distributed 3D location algorithm for mobile nodes in wireless sensor networks based on Monte Carlo. The algorithm uses the external cube to express the communication range and moving range of the node, forms the sampling area according to the maximum moving range of each slot of unknown node and the communication range of anchor node, and determines the sampling number according to the volume of anchor box. The random sampling points take the mean value as the estimated position of unknown nodes. The simulation results show that the influence of sampling number, time slot, anchor node ratio, node number, node moving speed, communication radius and the irregularity of communication range on the location error is analyzed. With small communication radius and anchor node ratio, the algorithm achieves ideal positioning accuracy and is superior to centroid algorithm, which effectively realizes the location of mobile nodes in three-dimensional environment. By introducing the effective sampling voting mechanism of neighbor nodes, this paper also studies a weighted Monte Carlo mobile node 3D location algorithm. The algorithm makes full use of the location information of neighbor nodes to vote the valid samples after screening and assigns larger weights to the samples closer to the real location. The weight value is used to reflect the influence of neighbor nodes on the estimated location to improve the location accuracy. The simulation results show that the localization accuracy of the algorithm is higher than that of the anchor box 3D algorithm and centroid algorithm through the analysis of the effects of different time slots, the proportion of anchor nodes, the number of nodes, the speed of node movement and the communication radius on the positioning error. The algorithm is robust, scalable and accurate. It is a distributed 3D localization algorithm without ranging. It can be used in mobile wireless sensor networks on a large scale.
【学位授予单位】：中南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5

【参考文献】