关键词: 物联网异构数据, 冗余特征, 汇聚节点, 最小二乘支持向量机, 聚类分析算法

Abstract: Aiming at the problems of slow speed and low accuracy of data fusion methods in the Internet of Things, in order to improve the effect of heterogeneous data fusion in the Internet of Things, we proposed a heterogeneous data fusion method of the Internet of Things based on intelligent optimization algorithm. Firstly, multinodes were used to collect the status data of monitoring objects, and the noise of data collected by each node was filtered to reduce the size of data initially and improve the quality of heterogeneous data of the Internet of Things. Secondly, clustering analysis algorithm was introduced to process the data of cluster head and eliminate the redundancy of data in the cluster. Finally, intelligent optimization algorithm was used to weigh the cluster head data in the aggregation node, and the contrast experiment was compared with other fusion methods under the same environment. The experimental results show that this method can effectively fuse heterogeneous data of the Internet of Things, and obtains highprecision results of heterogeneous data fusion of the Internet of Things. The heterogeneous data fusion of the Internet of Things has fewer errors and faster speed, which improves the efficiency of data fusion of the Internet of Things.

Key words: heterogeneous data of Internet of Things, redundant feature, aggregation node, least squares support vector machine, clustering analysis algorithm