三峡库区水环境监测网络节点定位研究

发布时间：2018-03-05 22:32

本文选题：无线传感器网络　切入点：三峡库区　出处：《重庆大学》2014年硕士论文　论文类型：学位论文

【摘要】：举世瞩目的长江三峡水利工程，已在我国社会经济发展中发挥重要作用。三峡水库水环境安全问题一直倍受关注。而现有的监测设备、方法和手段已不能满足监测任务需求，迫切需要提高三峡库区水环境监测的能力。无线传感器网络具有低成本、微型化、低能耗、自组织等特点，可用于三峡库区水环境监测，便于实现对库区水环境状况进行连续、实时监测和预警。监测应用中，节点提供的监测数据必须带有位置信息，才能准确判断可能存在的污染源。为此，本文在充分利用现有成果的基础上，就应用于三峡库区水环境监测的大规模无线传感器网络节点的定位问题展开研究。根据三峡库区独特的地理分布特征，构建的大规模无线传感器网络呈现出具有明显带状分布特性，常出现C型和Y型等特殊区域分布网络，这给节点定位带来很大困难。现有的定位算法很少考虑特殊复杂环境因素的影响，而且节点在完成定位后无法获知其位置的准确性，只能通过仿真实验数据进行评估。为此，本文在项目实际需求的基础上，结合三峡库区大规模无线传感器网络带状区域特性，将理论分析与仿真实验结合，提出了适用于三峡库区水环境监测网络节点的定位算法。论文主要工作如下： ①深入分析三峡库区传感器网络节点分布特性，并分别对复杂区域下的各向同性和各向异性网络进行研究。将博弈论用于传感器节点定位，建立了基于博弈论的传感器节点定位模型，并对模型求解。 ②给出了基于博弈论的定位算法的具体实现步骤；采用MATLAB仿真平台，验证该算法在复杂分布区域下传感器网络中节点的定位效果，并与另外几种算法相对比。实验证明该算法在各向同性和各向异性网络中均取得良好定位效果。 ③考虑在具体应用环境中，由于网络中采用某种算法进行节点定位的误差值难以获得，利用统计学中的抽样推断理论，，提出一种基于锚节点的定位误差估计方法，实现具体网络中节点定位误差估算。通过对多组网络进行实验，分析估算效果，结果表明该方法能够有效估计具体网络中的节点定位误差。
[Abstract]:The three Gorges Water Conservancy Project, which attracts worldwide attention, has played an important role in the social and economic development of our country. The problem of water environment safety of the three Gorges Reservoir has been paid close attention to. However, the existing monitoring equipment, methods and means can no longer meet the needs of the monitoring task. It is urgent to improve the ability of water environment monitoring in the three Gorges Reservoir area. Wireless sensor network has the characteristics of low cost, miniaturization, low energy consumption, self-organization, etc. It can be used for monitoring the water environment of the three Gorges Reservoir area, and it is convenient to carry out continuous water environment monitoring in the three Gorges Reservoir area. Real-time monitoring and early warning. In monitoring applications, monitoring data provided by nodes must have location information in order to accurately determine possible sources of pollution. The localization of large scale wireless sensor network nodes used in water environment monitoring in the three Gorges Reservoir area is studied. According to the unique geographical distribution characteristics of the three Gorges Reservoir area, the large scale wireless sensor networks have obvious zonal distribution characteristics, such as C type and Y type special regional distribution networks. This brings great difficulties to node location. Existing localization algorithms seldom consider the influence of special and complex environmental factors, and the node can not get the accuracy of its location after the location has been completed, so it can only be evaluated by simulation experimental data. On the basis of the actual demand of the project, this paper combines the theoretical analysis with the simulation experiment, combined with the zonal characteristics of the large-scale wireless sensor network in the three Gorges Reservoir area. This paper presents an algorithm for locating nodes in the water environment monitoring network of the three Gorges Reservoir area. The main work of this paper is as follows:. 1. The distribution characteristics of sensor network nodes in the three Gorges Reservoir area are analyzed, and the isotropic and anisotropic networks in complex regions are studied respectively. The game theory is used to locate sensor nodes. A sensor node location model based on game theory is established and solved. (2) the implementation steps of the localization algorithm based on game theory are given, and the localization effect of the algorithm in the sensor network is verified by using the MATLAB simulation platform. Compared with other algorithms, the experimental results show that the proposed algorithm achieves good localization performance in both isotropic and anisotropic networks. Considering that in the specific application environment, because it is difficult to obtain the error value of node location using some algorithm in the network, using the sampling inference theory in statistics, a localization error estimation method based on anchor node is proposed. The estimation of node location error in a specific network is realized. The results show that the proposed method can effectively estimate the node location error in a specific network by means of experiments on multiple networks and analysis of the estimation results.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TV697.2;TN929.5;TP212.9

【参考文献】