基于混支持度和粗糙集的异构数据融合研究

发布时间：2018-07-26 11:57

【摘要】：随着物联网和智慧城市的快速发展,异构数据融合技术越来越受到大家的关注,无论是在飞行目标跟踪、路径规划、环境检测,还是决策支持和智能交通等领域都显示出了巨大活力。异构数据融合技术可以充分利用多源信息间的互补,来更加全面、真实的反映感知对象的状态信息,目前已成为物联网的核心技术之一。本文对无线传感网络中的异构数据融合技术进行了比较深入、系统的研究,并针对异构数据融合模型、融合精度以及异常数据处理方面存在的问题,提出了相应的解决方案。通过算法比较和实验分析,证明了算法的有效性和可行性。首先,对实际WSN(Wireless Sensor Network)中测得的感知数据,从数据融合的角度分析感知数据的时空特性以及异常数据对融合结果的影响。在此基础上,提出了灰度预测的异常数据修正算法。同时,从感知数据的三维结构出发,将无线传感网络数据融合分割成同构数据融合和异构数据融合两个部分,并提出了异构数据的层次融合模型。其次,针对异构数据的层次融合模型,提出了一种基于混支持度和粗糙集的异构数据融合算法。该算法在同、异构数据类型的基础上,分别提出了基于混支持度的同构数据融合算法和基于粗糙集的异构数据融合算法,降低了异构数据的融合复杂性。最后,利用从实验环境中得到的实际测量数据,对本文提出的异构数据融合算法进行仿真实验,从融合精度和正确识别率两方面对算法的性能进行评估。实验结果表明,本文提出的基于混支持度和粗糙集的异构数据融合算法,在提高融合精度的同时,也提高了决策支持的准确性。
[Abstract]:With the rapid development of the Internet of things and smart cities, heterogeneous data fusion technology has attracted more and more attention. It has shown great vitality in the fields of flight target tracking, path planning, environment detection, decision support and intelligent transportation. Heterogeneous data fusion technology can make full use of the complementarity of multi source information. It has become one of the core technologies of the Internet of things, which is a more comprehensive and real reflection of the state information of the perceived object. This paper makes a thorough and systematic study of the heterogeneous data fusion technology in the wireless sensor network, and puts forward the problems in the heterogeneous data fusion model, the fusion precision and the abnormal data processing. According to the algorithm comparison and experimental analysis, the effectiveness and feasibility of the algorithm are proved. First, the spatial and temporal characteristics of the perceived data and the effect of abnormal data on the fusion results are analyzed from the point of view of the actual WSN (Wireless Sensor Network). On this basis, the gray level preview is proposed. At the same time, the data fusion of wireless sensor networks is divided into two parts: isomorphic data fusion and heterogeneous data fusion, and a hierarchical fusion model of heterogeneous data is proposed. Secondly, based on the heterogeneous data layer fusion model, a kind of mixed support degree is proposed. On the basis of the same and heterogeneous data types, this algorithm proposes an isomorphic data fusion algorithm based on mixed support and a heterogeneous data fusion algorithm based on rough sets, which reduces the complexity of heterogeneous data fusion. Finally, the actual measurement data obtained from the experimental environment are used in this paper. The performance of the algorithm is evaluated from the two aspects of the fusion precision and the correct recognition rate. The experimental results show that the heterogeneous data fusion algorithm based on the mixed support degree and the rough set proposed in this paper improves the accuracy of the fusion and improves the accuracy of the decision support.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP202;TN929.5;TP212.9

【参考文献】