无线传感器网络中数据融合算法的研究
发布时间：2018-07-18 13:19
【摘要】：无线传感器网络在军事、医疗、环境等诸多领域有着广阔的应用前景，因而受到越来越多的关注。无线传感器网络中节点能量受限，因此延长网络的生命周期成为首要考虑的问题。在网络中，节点的能量主要消耗在数据采集、处理和传输三方面，其中传输消耗得最多；而数据融合正是一种减少数据通信量的技术，在节能方面效果显著，现已经成为无线传感器网络的一个研究热点。 本文从现有的路由协议和数据融合算法着手研究，，提出了基于能量优化的可靠数据融合改进算法（EORDA）。（1）针对分层路由协议中簇头节点负载过重的问题，在簇的建立阶段，运用自适应k均值算法分簇，并综合考虑节点的位置、剩余能量及其可靠性选择簇头；在通信阶段，簇头将融合后的数据沿着蚁群算法找到的最优路径以多跳方式发送给基站；从而有效解决了均匀分簇以及簇头灵活选择通信方式的问题。（2）针对现有融合算法存在的容错性及通用性不强等问题，提出先依照节点信誉清洗原始数据，再运用核主成分分析法提取数据的主要信息，然后发送给基站进行决策分析，从而在确保数据融合可靠性的基础上大幅降低了数据通信量。 仿真结果表明：在五种网络场景中，该算法在均衡网络能耗方面均取得了良好效果，能使网络生命周期延长至少24%；在四种融合方案中，该算法不仅使网络中数据报个数大幅减少，而且使数据融合精度达到98%左右。
[Abstract]:Wireless sensor networks (WSN) have wide application prospects in military, medical, environmental and other fields, so more and more attention has been paid to wireless sensor networks. In wireless sensor networks (WSN), the energy of nodes is limited, so prolonging the lifetime of wireless sensor networks becomes the most important consideration. In the network, the energy of nodes is mainly consumed in three aspects: data acquisition, processing and transmission, among which transmission consumption is the most, and data fusion is a technology to reduce data traffic, which has remarkable effect in energy saving. Now it has become a research hotspot in wireless sensor networks. Based on the existing routing protocols and data fusion algorithms, an improved reliable data fusion algorithm based on energy optimization (Eorda). (1) is proposed in this paper, which aims at the problem of overloaded cluster head nodes in hierarchical routing protocols. The adaptive k-means algorithm is used to cluster and consider the location of nodes, residual energy and its reliability to select cluster heads. In the communication stage, the fused data is sent to the base station in a multi-hop way along the optimal path found by ant colony algorithm. Thus, the problem of uniform clustering and flexible selection of communication mode by cluster heads is solved effectively. (2) aiming at the problems of fault tolerance and universality of existing fusion algorithms, it is proposed that the original data should be washed according to the reputation of nodes first. Then the kernel principal component analysis (KPCA) is used to extract the main information of the data, and then send it to the base station for decision analysis, thus greatly reducing the data traffic on the basis of ensuring the reliability of the data fusion. The simulation results show that the proposed algorithm achieves good results in balancing network energy consumption in five network scenarios, and can prolong the network life cycle by at least 24%. The algorithm not only reduces the number of Datagram in the network, but also makes the data fusion accuracy about 98%.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5
[Abstract]:Wireless sensor networks (WSN) have wide application prospects in military, medical, environmental and other fields, so more and more attention has been paid to wireless sensor networks. In wireless sensor networks (WSN), the energy of nodes is limited, so prolonging the lifetime of wireless sensor networks becomes the most important consideration. In the network, the energy of nodes is mainly consumed in three aspects: data acquisition, processing and transmission, among which transmission consumption is the most, and data fusion is a technology to reduce data traffic, which has remarkable effect in energy saving. Now it has become a research hotspot in wireless sensor networks. Based on the existing routing protocols and data fusion algorithms, an improved reliable data fusion algorithm based on energy optimization (Eorda). (1) is proposed in this paper, which aims at the problem of overloaded cluster head nodes in hierarchical routing protocols. The adaptive k-means algorithm is used to cluster and consider the location of nodes, residual energy and its reliability to select cluster heads. In the communication stage, the fused data is sent to the base station in a multi-hop way along the optimal path found by ant colony algorithm. Thus, the problem of uniform clustering and flexible selection of communication mode by cluster heads is solved effectively. (2) aiming at the problems of fault tolerance and universality of existing fusion algorithms, it is proposed that the original data should be washed according to the reputation of nodes first. Then the kernel principal component analysis (KPCA) is used to extract the main information of the data, and then send it to the base station for decision analysis, thus greatly reducing the data traffic on the basis of ensuring the reliability of the data fusion. The simulation results show that the proposed algorithm achieves good results in balancing network energy consumption in five network scenarios, and can prolong the network life cycle by at least 24%. The algorithm not only reduces the number of Datagram in the network, but also makes the data fusion accuracy about 98%.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5
【参考文献】
相关期刊论文 前10条
1 孙其博;刘杰;黎
本文编号：2132021
本文链接：https://www.wllwen.com/kejilunwen/wltx/2132021.html
