矿井中无线传感器网络数据融合技术研究

发布时间：2018-03-15 12:47

本文选题：矿井　切入点：无线传感器网络　出处：《河南理工大学》2014年硕士论文　论文类型：学位论文

【摘要】：目前矿井安全事故频发,如何建立有效的监控网络成为煤矿企业面临的难题。无线传感器网络能够适应复杂多变的矿井环境,提高了监控网络的可靠性。为确保有效覆盖和节点之间的连通性,同一区域往往同时处于多个节点的感应范围内,势必造成大量数据冗余,需进行数据融合。数据融合能够降低网络总能耗,因此其成为WSN领域研究的热点问题。本课题分析现有融合算法存在的优缺点,从矿井监控的角度提出了一种WSN数据融合技术。主要从网络能耗、传输时延和数据精确性三个方面进行WSN数据融合模型的研究,相关工作如下:(1)为了降低网络的总能耗,提出一种基于分簇的WSN数据融合算法。分簇过程中采用了一种平均能耗最优的均匀分簇算法,在簇头选择时,考虑了节点剩余能量、传输距离和当选次数对节点当选簇头的概率的影响;数据传输过程中提出一种估计机制的融合算法,将相邻两个采集周期数据的变化量的绝对值与设定阈值进行比较,去除阈值内的冗余数据,大大降低了网络总能耗;最后分析了阈值大小与网络总能耗的关系,并且当网络总能耗最小时得到最佳阈值。(2)鉴于估计机制的数据融合算法降低网络能耗的同时会增加传输时延,降低数据精确性,本文提出一种均衡时延和数据精确性的自适应融合算法。通过引入博弈理论和自适应融合因子,当矿井环境中丢包率动态变化时,簇头自适应选择最佳融合因子使网络总效益函数值最大,最佳融合因子总体上随着丢包率的增大而增大,这样就有效均衡了传输时延和数据的精确性。(3)均匀分簇算法导致不同监测区域的簇的能耗速率不均匀,本文提出一种均衡能耗速率的非均匀分簇融合算法。在非均匀分簇阶段,提出了节点“簇半径”的概念,使其随着监测区域与sink距离的增加而增加;在数据融合阶段,当丢包率随着簇半径动态变化时,在各个区域的簇内自适应调整融合因子的大小,并且融合因子的大小随着簇半径的增加而减小,减少了远离sink节点区域的数据量,均衡了网络能耗速率,同时降低了网络传输时延。
[Abstract]:At present, mine safety accidents occur frequently. How to establish an effective monitoring network becomes a difficult problem for coal mining enterprises. Wireless sensor network can adapt to the complex and changeable mine environment. In order to ensure effective coverage and connectivity between nodes, the same area is often in the inductive range of multiple nodes at the same time, which will inevitably result in a large amount of data redundancy. Data fusion is needed to reduce the total network energy consumption, so it has become a hot issue in the field of WSN. This paper analyzes the advantages and disadvantages of existing fusion algorithms. This paper puts forward a WSN data fusion technology from the angle of mine monitoring. It mainly studies the WSN data fusion model from three aspects of network energy consumption, transmission delay and data accuracy. The related work is as follows: 1) in order to reduce the total energy consumption of the network, A clustering based WSN data fusion algorithm is proposed, in which a uniform clustering algorithm with optimal average energy consumption is used, and the residual energy of the nodes is considered in the selection of cluster heads. In the process of data transmission, a fusion algorithm of estimation mechanism is proposed, which compares the absolute values of the changes of two adjacent data acquisition periods with the set threshold. The total energy consumption of the network is greatly reduced by removing the redundant data in the threshold value. Finally, the relationship between the threshold value and the total energy consumption of the network is analyzed. And when the total energy consumption of the network is minimum, the optimal threshold is obtained.) since the data fusion algorithm of the estimation mechanism reduces the energy consumption of the network, it will increase the transmission delay and reduce the accuracy of the data. In this paper, an adaptive fusion algorithm for equalizing delay and data accuracy is proposed. By introducing game theory and adaptive fusion factor, when packet loss rate changes dynamically in mine environment, The cluster head adaptively selects the best fusion factor to make the total benefit function of the network maximum, and the optimal fusion factor increases with the increase of packet loss rate. In this way, the transmission delay and the accuracy of the data are effectively equalized. The uniform clustering algorithm leads to the uneven energy consumption rate in different monitoring regions. In this paper, a non-uniform clustering algorithm for equalizing the energy consumption rate is proposed. The concept of node "cluster radius" is proposed, which increases with the increase of the distance between the monitoring area and the sink, and adaptively adjusts the fusion factor in each region when the packet loss rate changes dynamically with the cluster radius in the data fusion stage. And the size of fusion factor decreases with the increase of cluster radius, reducing the amount of data far away from the sink node area, equalizing the network energy consumption rate, and reducing the network transmission delay.
【学位授予单位】：河南理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TD65;TN929.5;TP212.9

【相似文献】