arXiv subjects

Dmytro Tymoshchuk

Publications and source records attributed to Dmytro Tymoshchuk.

6 recordsLinked to original sources

arXiv2025

Comparison of feature extraction tools for network traffic data

The comparison analysis of the most popular tools to extract features from network traffic is conducted in this paper. Feature extraction plays a crucial role in Intrusion Detection Systems (IDS) because it helps to transform huge raw network data into meaningful and manageable features for analysis and detection of malicious activities. The good choice of feature extraction tool is an essential step in construction of Artificial Intelligence-based Intrusion Detection Systems (AI-IDS), which can help to enhance the efficiency, accuracy, and scalability of such systems.

cs.CR↗

arXiv2025

Modelling of automotive steel fatigue lifetime by machine learning method

In the current study, the fatigue life of QSTE340TM steel was modelled using a machine learning method, namely, a neural network. This problem was solved by a Multi-Layer Perceptron (MLP) neural network with a 3-75-1 architecture, which allows the prediction of the crack length based on the number of load cycles N, the stress ratio R, and the overload ratio Rol. The proposed model showed high accuracy, with mean absolute percentage error (MAPE) ranging from 0.02% to 4.59% for different R and Rol. The neural network effectively reveals the nonlinear relationships between input parameters and fatigue crack growth, providing reliable predictions for different loading conditions.

cs.LG↗

arXiv2025

Automated monitoring of bee colony movement in the hive during winter season

In this study, we have experimentally modelled the movement of a bee colony in a hive during the winter season and developed a monitoring system that allows tracking the movement of the bee colony and honey consumption. The monitoring system consists of four load cells connected to the RP2040 controller based on the Raspberry Pi Pico board, from which data is transmitted via the MQTT protocol to the Raspberry Pi 5 microcomputer via a Wi-Fi network. The processed data from the Raspberry Pi 5 is recorded in a MySQL database. The algorithm for finding the location of the bee colony in the hive works correctly, the trajectory of movement based on the data from the sensors repeats the physical movement in the experiment, which is an imitation of the movement of the bee colony in real conditions. The proposed monitoring system provides continuous observation of the bee colony without adversely affecting its natural activities and can be integrated with various wireless data networks. This is a promising tool for improving the efficiency of beekeeping and maintaining the health of bee colonies.

eess.SY↗

arXiv2025

Using hypervisors to create a cyber polygon

Cyber polygon used to train cybersecurity professionals, test new security technologies and simulate attacks play an important role in ensuring cybersecurity. The creation of such training grounds is based on the use of hypervisors, which allow efficient management of virtual machines, isolating operating systems and resources of a physical computer from virtual machines, ensuring a high level of security and stability. The paper analyses various aspects of using hypervisors in cyber polygons, including types of hypervisors, their main functions, and the specifics of their use in modelling cyber threats. The article shows the ability of hypervisors to increase the efficiency of hardware resources, create complex virtual environments for detailed modelling of network structures and simulation of real situations in cyberspace.

cs.CR↗

arXiv2025

Detection and classification of DDoS flooding attacks by machine learning method

This study focuses on a method for detecting and classifying distributed denial of service (DDoS) attacks, such as SYN Flooding, ACK Flooding, HTTP Flooding, and UDP Flooding, using neural networks. Machine learning, particularly neural networks, is highly effective in detecting malicious traffic. A dataset containing normal traffic and various DDoS attacks was used to train a neural network model with a 24-106-5 architecture. The model achieved high Accuracy (99.35%), Precision (99.32%), Recall (99.54%), and F-score (0.99) in the classification task. All major attack types were correctly identified. The model was also further tested in the lab using virtual infrastructures to generate normal and DDoS traffic. The results showed that the model can accurately classify attacks under near-real-world conditions, demonstrating 95.05% accuracy and balanced F-score scores for all attack types. This confirms that neural networks are an effective tool for detecting DDoS attacks in modern information security systems.

cs.CR↗

arXiv2024

Interactive cybersecurity training system based on simulation environments

Rapid progress in the development of information technology has led to a significant increase in the number and complexity of cyber threats. Traditional methods of cybersecurity training based on theoretical knowledge do not provide a sufficient level of practical skills to effectively counter real threats. The article explores the possibilities of integrating simulation environments into the cybersecurity training process as an effective approach to improving the quality of training. The article presents the architecture of a simulation environment based on a cluster of KVM hypervisors, which allows creating scalable and flexible platforms at minimal cost. The article describes the implementation of various scenarios using open source software tools such as pfSense, OPNsense, Security Onion, Kali Linux, Parrot Security OS, Ubuntu Linux, Oracle Linux, FreeBSD, and others, which create realistic conditions for practical training.

cs.CR↗