Scholarlink

<p>Data transmissions and uploading are becoming frequent in the industrial Internet of Things (IIoT) domain due to the benefits provided. Several organizations and consumers have benefited from the IIoT since the network facility supplied by this domain is extensive and can serve a large number of users at the same time. One important concern with this platform is the introduction of attacks that are designed to undermine the system's stability and earn financial rewards. To address the issue, this paper provides a blockchain (BC)-based security and attack detection system that employs a number of mechanisms. The data are initially collected, and the encryption phase is carried out using the blowfish algorithm (BFA). The adaptive wild horse optimization (AWHO) technique is used to determine the best key to encrypt the data. After encryption, BC technology is used to convert the data into blocks, and Merkle Tree Hashing Algorithm (MTHA) is used to protect the data. After conversion, the data is uploaded to the network, and the decryption process is done on the user side. Then, the attack classification phase is performed using the attention-based capsule bidirectional gated recurrent unit (CapsBi-GRU) model. The work is implemented in the Python tool, and the evaluations are carried out using the UNSW-NB15 and TON-IoT data sets. The obtained results demonstrate the superiority of the proposed method compared to other existing works.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Blockchain based deep learning model for the security in industrial internet of things

  • 期刊论文

  • 发表日期：

    2025年6月7日

查看全文
  • Tejeswara Kumar M,
  • N. V. R. Vikram G

摘要

Data transmissions and uploading are becoming frequent in the industrial Internet of Things (IIoT) domain due to the benefits provided. Several organizations and consumers have benefited from the IIoT since the network facility supplied by this domain is extensive and can serve a large number of users at the same time. One important concern with this platform is the introduction of attacks that are designed to undermine the system's stability and earn financial rewards. To address the issue, this paper provides a blockchain (BC)-based security and attack detection system that employs a number of mechanisms. The data are initially collected, and the encryption phase is carried out using the blowfish algorithm (BFA). The adaptive wild horse optimization (AWHO) technique is used to determine the best key to encrypt the data. After encryption, BC technology is used to convert the data into blocks, and Merkle Tree Hashing Algorithm (MTHA) is used to protect the data. After conversion, the data is uploaded to the network, and the decryption process is done on the user side. Then, the attack classification phase is performed using the attention-based capsule bidirectional gated recurrent unit (CapsBi-GRU) model. The work is implemented in the Python tool, and the evaluations are carried out using the UNSW-NB15 and TON-IoT data sets. The obtained results demonstrate the superiority of the proposed method compared to other existing works.