Abstract Industrial internet of things (IIoT) equipments undergoes a wide variety of quality checks before being accepted for on-site usage. Because of recent advances in machine learning& cloud-based computations, IIoT devices have sufficiently high accuracy & precision performance. Due to which, IIoT network designers focus more towards improving security, privacy and scaliability. A wide variety of models are available for security which uses cryptographic, key-exchange and blockchain implementations. Due to their application specific nature, these models have low scalability, which limits their real-time usability. Hence to overcome this issue, current research proposes hybrid augmented blockchain & machine learning model which help to improve device scalability. This model utilizes side chains and machine learning models for improving QoS while maintaining high security in IIoT networks. It further proposes an interfacing method that is tested on multiple existing IIoT nodes, for enhancing their security & QoS performance. It is observed that the proposed model gives 18%better accuracy,11%better precision and 12%better AUC of attack detection when compared with NR2B, SMHGKA and DQNSBmodels. Thus, the proposed model is capable of high security, better QoS, and has superior scalability due to black-box approach for existing devices.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|