Classification Method of Blockchain and IoT Devices Based on LSTM

Pengyu Duan, Ruiguang Li, Liehuang Zhu*, Hao Yu

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

With the development of the technology, more and more blockchain devices and Internet of Things devices are deployed around us. In order to manage these devices more conveniently, it is necessary to classify the devices according to the flow rate. The existing classification methods of devices are divided into active detection and passive detection. Active detection needs to send messages to them. Passive detection only needs to get the flow information of the devices, and existing passive detection needs to do so. The methods are all based on complete flow. We propose a method based on the previous part of the flow, using LSTM and random forest algorithm to classify. The f1-score of our method is 0.89. Compared with the existing methods, our method improves by 8%–30%, and our method can classify devices more effectively.

Original languageEnglish
Title of host publicationBlockchain and Trustworthy Systems - 3rd International Conference, BlockSys 2021, Revised Selected Papers
EditorsHong-Ning Dai, Xuanzhe Liu, Daniel Xiapu Luo, Jiang Xiao, Xiangping Chen
PublisherSpringer Science and Business Media Deutschland GmbH
Pages355-367
Number of pages13
ISBN (Print)9789811679926
DOIs
Publication statusPublished - 2021
Event3rd International Conference on Blockchain and Trustworthy Systems, Blocksys 2021 - Guangzhou, China
Duration: 5 Aug 20216 Aug 2021

Publication series

NameCommunications in Computer and Information Science
Volume1490 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference3rd International Conference on Blockchain and Trustworthy Systems, Blocksys 2021
Country/TerritoryChina
CityGuangzhou
Period5/08/216/08/21

Keywords

  • Blockchain
  • Internet of things
  • LSTM
  • Network traffic
  • Random forest

Fingerprint

Dive into the research topics of 'Classification Method of Blockchain and IoT Devices Based on LSTM'. Together they form a unique fingerprint.

Cite this