Key points are not available for this paper at this time.
Abstract By providing ubiquitous connectivity, effective data analytics tools, and better decision support systems for improved market competitiveness, the industrial internet of things (IIoT) promises creative business models in different industrial domains. However, the conventional IIoT architecture can no longer provide adequate support for such an enormous device as the number of nodes, and network size increases. Therefore, several challenges, such as security, privacy, centralization, trust, and integrity prevents faster adaptation of IIoT applications. To address aforementioned challenges, we present a deep blockchain‐based trustworthy privacy‐preserving secured framework (DBTP2SF) for IIoT environment. This framework comprises of three modules, namely, trust management module, a two‐level privacy‐preservation module, and an anomaly detection module. In trustworthiness module, blockchain (BC)‐based address reputation system is proposed. In the two‐level privacy module a BC‐based enhanced proof of work technique is simultaneously applied with AutoEncoder, to transform cyber‐physical system data into a new reduced form that prevents inference and poisoning attacks. In the anomaly detection module, deep neural network is deployed. Finally, due to various limitations of current Cloud‐Fog infrastructure, we present a BC‐interplanetary file systems integrated Cloud‐Fog architecture, namely, BlockCloud and BlockFog to deploy proposed DBTP2SF framework in IIoT environment. The experiment is conducted using IIoT‐based realistic dataset, namely, ToN‐IoT. The performance analysis shows that the proposed approach outperforms using transformed dataset over peer privacy‐preserving intrusion detection strategies, and has obtained accuracy of 98.97%, and detection rate of 93.87%. Finally, we have shown the superiority of DBTP2SF framework over some of the recent state‐of‐art techniques in both non‐BC and BC‐based IIoT system.
Kumar et al. (Thu,) studied this question.