UDC 004.738.5
IMPLEMENTATION OF ADAPTIVE MODELS AND ALGORITHMS TO ENSURE SAFETY AND OPTIMIZE INTERACTION IN INDUSTRIAL INTERNET-OF-THINGS NETWORKS
M. S. Poborueva, Postgraduate Student, RSREU, Ryazan, Russia
orcid.org/0009-0005-6270-3152, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
O. A. Bodrov, Associate Professor, PhD in Technical Sciences, RSREU, Ryazan, Russia
orcid.org/0009-0005-7225-6704, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The task of developing adaptive mathematical models and algorithms for ensuring cyber-physical secu rity and optimizing the interaction of intelligent objects in industrial Internet of Things (IIoT) networks is considered. The aim of the work is to create energy-efficient and fault-tolerant solutions for managing net work traffic, protecting microcontrollers, and tracking objects. An adaptive traffic control model based on graph neural networks (GNN) has been developed, which reduces delays by 15-20% and power consumption by 10-15% compared to AODV and RPL protocols, and has been tested in NS-3 environment. Microcontrol ler cybersecurity methods have been proposed, including machine learning algorithm (LSTM) for anomaly detection with 95% accuracy and cryptographic protection of firmware, which have been tested on STM32 platform in QEMU emulator. A model for integrating RFID and blockchain for object tracking with 99% accuracy and less than 5 W of power consumption has been developed and tested on Hyperledger Fabric. An algorithm for comprehensive assessment of IIoT performance has been created, combining security, power consumption, and performance metrics with 90% prediction accuracy. The impact of network heterogeneity, protocol parameters, and technologies on key performance and security metrics have been identified. The practical significance lies in the ability to select optimal technologies and improve the security and efficiency of IIoT systems.
Key words: : industrial Internet of Things (IIoT), adaptive mathematical models, cyber-physical security, graph neural networks (GNN), machine learning (LSTM), routing, energy efficiency, microcontroller cyber security, cryptography, RFID, blockchain, Hyperledger Fabric, comprehensive assessment, protocol classi fication, NS-3, STM32, QEMU.