19 December 2024 in Smart technologies: Industrial and Civil Engineering
The work considers modern approaches to the implementation of IoT technologies for effective management of urban infrastructure. The main focus is on the analysis of network architectures used for data collection and transmission in smart city systems. Various network construction models are compared, including single-hop and multi-hop networks, as well as WSN topologies. The work focuses on the advantages and disadvantages of these networks in the context of energy consumption, scalability and stability. Special attention is paid to hierarchical models that allow to reduce energy costs and increase network efficiency. The application of IoT in key areas of urban infrastructure, such as transport, energy and water supply, is analyzed. The issue of cyber security of IoT systems, including authentication, data encryption and ensuring communication reliability, is also discussed in detail. The role of CPS and sensor nodes that provide real-time monitoring and control of urban resources is discussed separately. Mathematical models for traffic optimization in IoT networks are considered, in particular, the use of queuing theory to ensure QoS. Conclusions have been drawn regarding the effectiveness of IoT implementation for urban resource management, which allows to increase the reliability of urban infrastructure and improve the living conditions of residents. The use of microservice software in IoT systems is substantiated and the directions of scientific research aimed at and its implementation in IoT-systems of management of infrastructure objects. The study opens up new perspectives for the application of IoT in smart cities and indicates the need for further study of security and energy conservation issues in such systems.
14 May 2025 in Smart technologies: Industrial and Civil Engineering
The paper analyzes the process of identifying incoming traffic in 5G/IMT-2020 networks, built on the Ultra-reliable and low latency communications technology, identifies its features and research directions to improve the efficiency of traffic identification taking into account the requirements for the functioning of the considered network. To solve the problem of identifying traffic in the 5G/IMT-2020 network, the paper develops and presents an appropriate methodology. The specified methodology includes the formation of metadata arrays of the incoming stream, their modification into a set of training data, the formation of a training software and hardware complex and the development of the neural network structure, the training process of the neural network and its implementation in the process of identifying traffic in 5G/IMT-2020 telecommunication networks. Evaluation of the results of the training process of the proposed neural network and verification of its operation on test data sets in the trained state showed that the neural network presented in the work is able to monitor and identify traffic generated from Internet of Things services with a probability of up to 99.7%. In the process of monitoring and identifying traffic from two or more services, this probability may decrease, but is within the acceptable limits of 80-90%.
1 January 2025 in Collection of scientific works of the Military Institute of Kyiv National Taras Shevchenko University
The article considers approaches to ensuring high-quality service of Internet of Things (IoT) services in 5G networks, taking into account modern challenges in terms of scalability, variable topology and security threats. Considerable attention is paid to modeling IoT traffic, building mathematical service models and assessing parameters that affect the level of reliability and timeliness of data delivery. During the study, the specifics of the structure of IoT networks were analyzed, key traffic metrics were identified and a mathematical model was built based on the classical Poisson model, as well as its extensions MMPP and BMAP to adequately reflect the variable and clustered traffic structure in IoT networks. Clustering was performed using the k-means algorithm to profile normal node behavior, which made it possible to detect anomalies in real time. A formal risk assessment model was proposed that takes into account deviations of current metrics from a typical cluster and changes in intensity. An adaptive incident response mechanism with threshold values that determine the level of intervention from limiting activity to completely blocking the device. For verification, DDoS, device substitution, and stealth attack scenarios were simulated. The effectiveness of the detection algorithm was assessed using the metrics precision, recall, F1-score, and response time. The results showed that the proposed methodology provides high anomaly detection accuracy (F1-score = 92.5%) with a response time of up to 80 ms, which makes it suitable for use in real IoT systems. The proposed approach allows for increased network reliability and security without significantly increasing computational costs. The findings confirm the feasibility of implementing intelligent monitoring systems based on clustering and risk models in the 5G-IoT infrastructure.