Blockchain enable IoT using deep reinforcement learning: A novel architecture to ensure security of data sharing and storage

Document Type

Conference Proceeding

Publication Title

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volume

13340 LNCS

First Page

586

Last Page

597

Publisher

Springer

School

School of Engineering

RAS ID

51841

Comments

Bai, X., Tu, S., Waqas, M., Wu, A., Zhang, Y., & Yang, Y. (2022). Blockchain enable IoT using deep reinforcement learning: A novel architecture to ensure security of data sharing and storage. In International Conference on Artificial Intelligence and Security (pp. 586-597). Springer. https://doi.org/10.1007/978-3-031-06791-4_46

Abstract

With the continuous development of the Internet of Things, more and more social sectors and smart devices are connected to the Internet of Things. This has led to a spurt of data growth and posed challenges to data security. To solve this problem, the Internet of Things needs a more secure and efficient storage method. Nowadays, one of the most important creative technological advancements that plays a significant role in the professional world is blockchain technology. And Both academia and industry attach great importance to the research of blockchain application technology. Some scholars believe that the blockchain itself is a secure distributed database. So, Blockchain is also considered a safe way to store data. In this paper, we introduce blockchain into the Internet of Things to ensure the security of Internet of Things data. At the same time, we have solved the problem of quantifying the degree of blockchain decentralization, which provides conditions for system optimization. After that, we proposed a system optimization model based on deep reinforcement learning to dynamically adjust system parameters. The simulation results show that the decentralization of the blockchain and the security of the system are guaranteed, and the throughput of the system has been improved.

DOI

10.1007/978-3-031-06791-4_46

Access Rights

subscription content

Share

 
COinS