无线传感器网络时空相关性数据融合算法研究

发布时间：2018-05-10 16:49

本文选题：无线传感器网络 + 数据融合　；参考：《重庆大学》2014年硕士论文

【摘要】：作为物联网的底层技术支撑，无线传感器网络采集和传输监测区域的各种信息，，为军事国防、远程医疗和环境监测等应用提供实时可靠的数据，是物联网的关键信息传输技术。但是，无线传感器网络节点分布密集并且采样频繁。感知数据之间的时空相关性造成冗余数据。大量冗余数据的传输对无线传感器网络有限的能量、存储能力和网络带宽带来了巨大的挑战。数据融合能有效地减少冗余数据，提高数据收集效率和准确性。因此，无线传感器网络数据融合研究具有重要的学术意义和工程价值。本文深入地研究了典型的数据融合算法的原理、特点和性能指标，结合单个无线传感器网络感知数据的时间相关性和多个无线传感器网络感知数据的空间相关性，提出时空相关性数据融合算法的架构，设计了两种高效的数据融合算法。本文主要的研究内容有如下几点： ①针对单个无线传感器网络节点感知数据冗余度高的问题，本文提出了一种基于时间相关性的数据融合算法。在分段一元线性回归模型的基础上，通过对无线传感器网络感知数据的时间序列进行分析，建立预测模型，并根据该模型的各个参数和给定的误差，自适应地调整下一个采集时间，并动态地优化回归模型。仿真和实验结果证明，针对不同的数据变化率，该算法均能减少数据采集量和传输量，满足数据精度。 ②针对多个无线传感器网络节点感知数据冗余度高的问题，本文提出一种基于空间相关性的数据融合算法。该算法将监测区域根据无线传感器网络感知数据的空间相关性的强弱程度划分为若干个相关区域。每个相关区域根据无线传感器网络节点的剩余能量，选取一个代表节点。多个相关区域根据代表节点的剩余能量和到汇聚节点的距离，选取一个簇头节点。代表节点将相关区域的感知数据融合后传递给簇头节点。簇头节点将数据通过无线方式转发给汇聚节点。仿真和实验结果证明，该算法使得代表节点和簇头节点分布均匀，数据传输量小，数据精度高。该算法具有很好的能效性。
[Abstract]:As the underlying technology support of the Internet of things, wireless sensor network (WSN) collects and transmits various kinds of information in the monitoring area, which provides real-time and reliable data for military defense, telemedicine and environmental monitoring applications. It is the key information transmission technology in the Internet of things. However, wireless sensor network nodes are densely distributed and frequently sampled. The temporal and spatial correlation between perceptual data results in redundant data. The transmission of large amounts of redundant data brings great challenges to the limited energy, storage capacity and bandwidth of wireless sensor networks. Data fusion can effectively reduce redundant data and improve the efficiency and accuracy of data collection. Therefore, the research of wireless sensor network data fusion has important academic significance and engineering value. In this paper, the principle, characteristics and performance index of typical data fusion algorithm are deeply studied, and the temporal correlation of sensing data of single wireless sensor network and the spatial correlation of sensing data of multiple wireless sensor networks are combined. The framework of spatio-temporal correlation data fusion algorithm is proposed, and two efficient data fusion algorithms are designed. The main contents of this paper are as follows: In order to solve the problem of high data redundancy in a single wireless sensor network node, this paper proposes a data fusion algorithm based on time correlation. Based on the piecewise univariate linear regression model, the prediction model is established by analyzing the time series of sensor network perceptual data, and according to each parameter and given error of the model, The next acquisition time is adjusted adaptively and the regression model is dynamically optimized. The simulation and experimental results show that the algorithm can reduce the amount of data acquisition and transmission and meet the accuracy of the data. In order to solve the problem of high data redundancy of multiple nodes in wireless sensor networks, this paper proposes a data fusion algorithm based on spatial correlation. The algorithm divides the monitoring region into several related regions according to the spatial correlation degree of the sensor network perceptual data. According to the residual energy of wireless sensor network nodes, each related region selects a representative node. A cluster head node is selected according to the residual energy of the node and the distance from the node to the convergence node. The representative node fuses the perceptual data of the related region and passes it to the cluster head node. The cluster head node forwards the data to the convergent node by wireless way. The simulation and experimental results show that the algorithm makes the representative node and cluster head node distribute uniformly, the data transmission is small, and the data precision is high. The algorithm has good energy efficiency.
【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】：TN929.5;TP212.9

【参考文献】