Classification and Processing of Big Data in Sensor Network Based on Suffix Tree Clustering

International Journal of Online and Biomedical Engineering

Jun Tian¹  Lirong Huang¹ 

[1] Foshan Polytechnic;

关键词: sensor network;    big data;    storage system;    suffix tree;    clustering;   

DOI  :  10.3991/ijoe.v15i01.9785

来源: DOAJ

【 摘 要 】

Aiming at the perception data acquired by the widely used, fast-developing but still not perfect wireless sensor network system, a relatively complete and universal system for the collection, transmission, storage and cluster analysis of perception data is designed. Perception data is spliced and compressed at the node and reconstructed at the base station, the problem of the acquisition of perception data and energy consumption of transmission is optimized, the distributed storage system is established, and the data reading mechanism and data storage architecture are designed accordingly.The data acquisition protocol and the traditional protocol, the storage system itself and the Oracle database system, and Standard Deviation and Eigensystem Realization Algorithm are respectively adopted for comparison test.Based on Standard Deviation algorithm, the operation of suffix tree clustering is carried out, and the general steps of suffix tree clustering are studied and the structure of perception data and the characteristics of storage are adapted, and the data classification operation based on suffix tree clustering is completed. The results show that proposed Standard Deviationalgorithm algorithm not only inherits the efficiency of the classical algorithm for processing big data, but also has obvious effect on large-scale discrete data processing, and the efficiency is obviously improved compared with the traditional method.

【 授权许可】

Unknown