GRADE: Deep learning and garlic routing-based secure data sharing framework for IIoT beyond 5G

Abstract

The rise of automation with Machine-Type Communication (MTC) holds great potential in developing Industrial Internet of Things (IIoT)-based applications such as smart cities, Intelligent Transportation Systems (ITS), supply chains, and smart industries without any human intervention. However, MTC has to cope with significant security challenges due to heterogeneous data, public network connectivity, and inadequate security mechanism. To overcome the aforementioned issues, we have proposed a blockchain and garlic-routing-based secure data exchange framework, i.e., GRADE, which alleviates the security constraints and maintains the stable connection in MTC. First, the Long-Short-Term Memory (LSTM)-based Nadam optimizer efficiently predicts the class label, i.e., malicious and non-malicious, and forwards the non-malicious data requests of MTC to the Garlic Routing (GR) network. The GR network assigns a unique ElGamal encrypted session tag to each machine partaking in MTC. Then, an Advanced Encryption Standard (AES) is applied to encrypt the MTC data requests. Further, the Inter-Planetary File System (IPFS)-based blockchain is employed to store the machine's session tags, which increases the scalability of the proposed GRADE framework. Additionally, the proposed framework has utilized the indispensable benefits of the 6G network to enhance the network performance of MTC. Lastly, the proposed GRADE framework is evaluated against different performance metrics such as scalability, packet loss, accuracy, and compromised rate of the MTC data request. The results show that the GRADE framework outperforms the baseline methods in terms of accuracy, i.e., 98.9%, compromised rate, i.e., 18.5%, scalability, i.e., 47.2%, and packet loss ratio, i.e., 24.3%.

Keywords

• Garlic routing
• Blockchain
• I2P
• LSTM
• Artificial intelligence
• Onion routing