Deep Learning in Security of Internet of Things

【Author】 Li, Yuxi; Zuo, Yue; Song, Houbing; Lv, Zhihan

【Source】IEEE INTERNET OF THINGS JOURNAL

【影响因子】10.238

【Abstract】Internet-of-Things (IoT) technology is increasingly prominent in the current stage of social development. All walks of life have begun to implement the IoT integration technology, so as to strive to promote industrial modernization, intelligence, and digitalization. In this case, how to link high-risk network activities with entities has become the primary issue for promoting industrial development. However, at this stage, the security issues in the development of the IoT technology have contradictions that are difficult to resolve. According to this situation, how to make system defense intelligent and replace manual monitoring has become the future of the development of security architecture. This article combines existing security research to explore the possibility of deep learning (DL) in upgrading the IoT security architecture, discusses how the IoT can identify and respond to cyber attacks, and how to encrypt edge data transmission. Moreover, this article discusses security research in application fields, such as Industrial IoT, Internet of Vehicles, smart grid, smart home, and smart medical. Then, we summarized the areas that can be improved in future technological development, including sharing computing power through the edge network processing unit (NPU) central device and closely combining the environmental simulation model with the actual environment, as well as malicious code detection, intrusion detection, production safety, vulnerability detection, fault diagnosis, and blockchain technology.

【Keywords】Security; Internet of Things; Deep learning; Feature extraction; Blockchains; Satellites; Safety; Deep learning (DL); Internet of Things (IoT); neural network processing unit (NPU); security

【发表时间】2022 15-Nov

【收录时间】2022-11-21

【文献类型】实证数据

【主题类别】

区块链技术-协同技术-深度学习