基于物联网的数据融合算法研究
发布时间：2018-12-19 21:02
【摘要】：当今社会，各种各样的通信手段和技术不断的更新、发展，同时，网络的发展也从互联网转向了所有的设备和机器都可以互相连通的方向，实现NetworkEverything，也就是我们所说的物联网通信。物联网（Internet of Things,IOT）可以被称为是新一代信息技术重要的组成部分，它最终实了现物和物、物和人以及人和人之间的相互连通。 物联网中，节点分布非常广泛，而且数量巨大，在这种情况下，对各节点采集的数据进行必要的融合处理是很重要的。数据融合技术已经发展的很成熟，，但是由于物联网与其他网络相比有其本身的特点，所以当把已有的数据融合算法直接应用于物联网中时会产生很多的不适应因素，本篇论文主要研究的就是数据融合不适应物联网特点的方面，通过分析研究针对这些特点提出具体的改进方案。 物联网的特点之一是数据的海量性，数据融合算法直接应用于物联网时会因为物联网的海量数据而导致融合节点能量的大量消耗甚至节点的过早死亡，所以需要对物联网的数据融合算法进行改进，使其适应物联网的这一特点。另外，物联网对数据具有实时性的要求，而且对不同类型的数据实时性要求是不同的，如果融合之前数据是按照原路由方式进行传输，即使进行了正确的融合处理，得到的结果也不是物联网所需要的，即没有体现出物联网数据实时性不同这一特点。所以在改进融合算法之前还应该首先对数据的传输方式进行改进，为数据融合结果的准确性打好前提，这才能使得到的结果符合物联网的特点。 为了完成以上任务，本文主要做了两方面的工作。一、为了使待融合的数据符合物联网中对数据实时性要求不同的特点，论文选取了基础的LEACH算法作为原型进行改进，通过改进使得各类数据能够按照不同的时间长度进行传输，保证了融合的结果能够符合物联网的特点，为数据融合打下了良好的基础。二、针对物联网中数据海量性这一特点，论文以模糊算法为基础，通过阈值范围的设置，减少了除首次外的融合数据的数据量，然后引入了权重值的概念，使其更适应剩余数据的融合。最终使得数据融合可以更好的应用在物联网的海量数据环境下。 在论文最后，对以上提到的改进算法进行了仿真，仿真结果表明，数据融合基于物联网中两方面特点的改进算法都达到了预期的效果，即使得数据融合更好地适应了物联网的新环境。
[Abstract]:In today's society, various communication means and technologies are constantly updated and developed. At the same time, the development of the network has also shifted from the Internet to the direction in which all devices and machines can be connected to each other, so as to realize NetworkEverything,. This is what we call Internet of things communications. The Internet of things (Internet of Things,IOT) can be described as an important part of the new generation of information technology. In the Internet of things, the nodes are widely distributed and have a large number of nodes. In this case, it is very important to perform the necessary fusion processing of the data collected by each node. Data fusion technology has been developed very mature, but because the Internet of things has its own characteristics compared with other networks, when the existing data fusion algorithm is directly applied to the Internet of things, there will be a lot of unsuitable factors. In this paper, the main research is that data fusion is not suitable for the characteristics of the Internet of things, through the analysis of these characteristics to put forward a specific improvement scheme. One of the characteristics of the Internet of things is the magnanimity of the data. When the data fusion algorithm is directly applied to the Internet of things, it will lead to a large amount of energy consumption and even premature death of the nodes because of the mass data of the Internet of things. So we need to improve the data fusion algorithm of the Internet of things to adapt to this characteristic of the Internet of things. In addition, the Internet of things has real-time requirements for data, and different types of data real-time requirements are different, if the data before fusion is transmitted according to the original routing mode, even if the correct fusion processing, The results obtained are not required by the Internet of things, that is, it does not reflect the difference in real-time performance of the Internet of things data. Therefore, before improving the fusion algorithm, we should first improve the transmission mode of data, and make a good premise for the accuracy of the data fusion results, which can make the results accord with the characteristics of the Internet of things. In order to complete the above task, this paper mainly do two aspects of work. First, in order to make the data to be fused in accordance with the different characteristics of real-time data in the Internet of things, this paper selects the basic LEACH algorithm as the prototype to improve the data transmission according to different time length. It ensures that the fusion results can accord with the characteristics of the Internet of things, and lays a good foundation for data fusion. Secondly, aiming at the characteristic of data magnanimity in the Internet of things, based on fuzzy algorithm, this paper reduces the amount of data of fusion data except for the first time by setting the threshold range, and then introduces the concept of weight value. Make it more suitable for the fusion of the remaining data. Finally, data fusion can be better applied in the mass data environment of the Internet of things. At the end of the paper, the improved algorithm mentioned above is simulated. The simulation results show that the improved algorithm of data fusion based on the two aspects of the Internet of things has achieved the desired results. Even if data fusion is better adapted to the new environment of the Internet of things.
【学位授予单位】：辽宁大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP391.44;TN929.5
本文编号：2387448
[Abstract]:In today's society, various communication means and technologies are constantly updated and developed. At the same time, the development of the network has also shifted from the Internet to the direction in which all devices and machines can be connected to each other, so as to realize NetworkEverything,. This is what we call Internet of things communications. The Internet of things (Internet of Things,IOT) can be described as an important part of the new generation of information technology. In the Internet of things, the nodes are widely distributed and have a large number of nodes. In this case, it is very important to perform the necessary fusion processing of the data collected by each node. Data fusion technology has been developed very mature, but because the Internet of things has its own characteristics compared with other networks, when the existing data fusion algorithm is directly applied to the Internet of things, there will be a lot of unsuitable factors. In this paper, the main research is that data fusion is not suitable for the characteristics of the Internet of things, through the analysis of these characteristics to put forward a specific improvement scheme. One of the characteristics of the Internet of things is the magnanimity of the data. When the data fusion algorithm is directly applied to the Internet of things, it will lead to a large amount of energy consumption and even premature death of the nodes because of the mass data of the Internet of things. So we need to improve the data fusion algorithm of the Internet of things to adapt to this characteristic of the Internet of things. In addition, the Internet of things has real-time requirements for data, and different types of data real-time requirements are different, if the data before fusion is transmitted according to the original routing mode, even if the correct fusion processing, The results obtained are not required by the Internet of things, that is, it does not reflect the difference in real-time performance of the Internet of things data. Therefore, before improving the fusion algorithm, we should first improve the transmission mode of data, and make a good premise for the accuracy of the data fusion results, which can make the results accord with the characteristics of the Internet of things. In order to complete the above task, this paper mainly do two aspects of work. First, in order to make the data to be fused in accordance with the different characteristics of real-time data in the Internet of things, this paper selects the basic LEACH algorithm as the prototype to improve the data transmission according to different time length. It ensures that the fusion results can accord with the characteristics of the Internet of things, and lays a good foundation for data fusion. Secondly, aiming at the characteristic of data magnanimity in the Internet of things, based on fuzzy algorithm, this paper reduces the amount of data of fusion data except for the first time by setting the threshold range, and then introduces the concept of weight value. Make it more suitable for the fusion of the remaining data. Finally, data fusion can be better applied in the mass data environment of the Internet of things. At the end of the paper, the improved algorithm mentioned above is simulated. The simulation results show that the improved algorithm of data fusion based on the two aspects of the Internet of things has achieved the desired results. Even if data fusion is better adapted to the new environment of the Internet of things.
【学位授予单位】：辽宁大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP391.44;TN929.5
【参考文献】
相关期刊论文 前6条
1 黄漫国;樊尚春;郑德智;邢维巍;;多传感器数据融合技术研究进展[J];传感器与微系统;2010年03期
2 王刚,赵海,魏守智;基于威胁博弈理论的决策级融合模型[J];东北大学学报;2004年01期
3 刘娅;徐震;杨蕾;;基于物联网的无线传感器网络路由协议研究[J];光通信研究;2014年01期
4 童树鸿,沈毅,刘志言;数据融合的模糊模型和算法研究[J];哈尔滨工业大学学报;2002年01期
5 周开乐;丁帅;胡小建;;面向海量数据应用的物联网信息服务系统研究综述[J];计算机应用研究;2012年01期
6 陈闻杰;陈迅;高丽强;陈章龙;涂时亮;;无线传感器网络成簇算法研究[J];小型微型计算机系统;2008年02期
本文编号：2387448
本文链接：https://www.wllwen.com/kejilunwen/wltx/2387448.html
