Protection and detection of physical hazards for the intelligent system under industrial Internet of Things by neural network. Issue 2 (30th December 2021)

• Record Type: Journal Article
• Authors: Wang, Rongxia; Liu, Zhifang; He, Fen; Yang, Weihuang
• Other Names: Souri Alireza guestEditor.; Al‐Masri Eyhab guestEditor.; Kumari Saru guestEditor.
• Is Part Of: Security and privacy. Volume 6:Issue 2(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-30
• Subjects: intelligent system of the industrial IoT -- Kalman filtering algorithm -- Kalman‐BPNN algorithm -- neural network; Computer security -- Periodicals; Data protection -- Periodicals; Cyberterrorism -- Periodicals; 005.8
• DOI: 10.1002/spy2.207
• Languages: English
• ISSNs: 2475-6725

Abstract

Abstract: Aiming at detecting and avoiding many security risks brought by various intelligent equipment and sensors in industrial Internet of Things (IoT) intelligent system, the Kalman back propagation neural network (Kalman‐BPNN) algorithm is designed, based on the neural network and with the help of the unique advantages of Kalman filtering algorithm and back propagation neural network (BPNN) algorithm. Introductions are made on the architecture of IoT and the theory of system security, and descriptions are made on the principle and process of related algorithms in detail. The advantages of traditional algorithms are integrated to design a Kalman‐BPNN algorithm. The simulation experiment analysis is carried out, and the designed detection method is compared with the traditional detection method. The voltage test of the Kalman‐BPNN algorithm based on the neural network is no more than 4.1%, as the experiment shows. The true positive rate of Kalman‐BPNN algorithm is 91.121%, which is very close to the value of 1, indicating that the algorithm has a great probability to predict the normal voltage. The false positive rate is 9.341%, which is very close to 0, indicating that the algorithm has a small probability of false judgment.