May 2025
Integrating a DNS monitoring module into a cybersecurity architecture to enhance protection against spoofing and phishing attacks
2025, vol.17 , no.2, pp. 111-122
Article [2025-02-11]
This paper is considered to be based on previous research of the author devoted to the conceptual design of adaptive security systems architecture with a focus on the development of a cybersecurity module using Enterprise Architect. The aim of the paper is to present methods and means for conducting regular monitoring and analysis for the prevention of spoofing and phishing attacks. The theoretical framework of the study is supported by a methodology for testing and simulations using Simple DNS Plus, which contributes to detecting an irregular increase in the frequency of DNS (Domain Name System) request.
Yoana Ivanova. Integrating a DNS monitoring module into a cybersecurity architecture to enhance protection against spoofing and phishing attacks. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 111-122.
Enhanced route reliability algorithm for vehicular ad hoc networks
2025, vol.17 , no.2, pp. 101-110
Article [2025-02-10]
Vehicular Ad Hoc Networks (VANETs) play a crucial role in facilitating road communication. By transmitting real-time traffic information to nearby vehicles and infrastructure, VANETs help address various real-world challenges, enhancing traffic management and road safety. To achieve this a secure and reliable scheme is needed. Reliable communication in VANER struggles with many problems like varying speed of vehicles, dynamic vehicle movement etc. To address this, we propose the Clustering Based Enhanced Route Reliability Algorithm (ERRAV), which uses clustering technique to improve route stability. In this geographical area is divided into clusters, each cluster is managed by a Cluster Head (CH). CH is selected based on signal strength, mobility, and vehicle density, ensuring stable communication. After clustering Reliable routes are established and maintained through periodic link monitoring, route updates, and backup routes. Simulations of ERRAV show the enhancement of scalability by reducing congestion, optimizing route discovery, and adaptation to network changes. EERAV been compared to AODV, DSR, and OLSR and experimental results show the reduced routing overhead, improves packet delivery, and minimizes link failures, making it highly effective in mobile environments.
Wajid Ali, Shalini Z. Ninoria. Enhanced route reliability algorithm for vehicular ad hoc networks. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 101-110.
A Chaotic-Based Image Encryption System Integrated with Blockchain for Secure Data Transmission
2025, vol.17 , no.2, pp. 89-100
Article [2025-02-09]
In this research, we present an exploration of encryption cryptosystem that integrates chaotic systems and blockchain technology to protect the privacy of your information and ensuring that its trustworthy. The system applies Chua's system, a communally known chaotic model, besides a Bernoulli binary generator to make a robust chaotic sequence for encrypting images. Blockchain technology further strengthens security by maintaining the picture's integrity hash and decryption key inside the blockchain, to make sure the encryption keys are tamper-resistant and sustain an immutable record for secure access and recovery. Additionally, the encryption process manufactures about 1-million-bit stream, which successfully passes NIST tests, confirming the randomness and security of the bit stream and refine the cryptosystem's general robustness. This method offers a great degree of security for sensitive and medical image data, protecting confidentiality and integrity via the incorporation of chaotic encryption, blockchain-based key management, and conforming random bit streams. The system is ideally designed to safeguard the image transmission in fields such as telemedicine and other healthcare-related applications.
Ahmed S. Alshammari. A Chaotic-Based Image Encryption System Integrated with Blockchain for Secure Data Transmission. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 89-100.
AI-driven resources allocation strategies for enhanced quality of service in 5G networks
2025, vol.17 , no.2, pp. 77-88
Article [2025-02-08]
This study explores AI-driven strategies for resource allocation in 5G networks to optimize Quality of Service (QoS). Given the growing demand for high-speed, low-latency connectivity, efficient bandwidth distribution is critical. The research analyses a dataset containing various application types, signal strengths, latencies, and bandwidth requirements to assess the efficiency of current resource allocation mechanisms. The study examines the impact of dynamic network conditions on resource distribution and proposes AI-based solutions, including neural networks and reinforcement learning, to enhance decision-making. Experimental results show that Gradient Boosting reduces RMSE by 18% compared to traditional heuristic-based allocation, while reinforcement learning models exhibit superior adaptability. The proposed framework leverages machine learning to optimize bandwidth allocation, reduce congestion, and enhance user experience in next-generation mobile networks.
Abdullah Havolli, Majlinda Fetaji. AI-driven resources allocation strategies for enhanced quality of service in 5G networks. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 77-88.
Semantic knowledge models of non-crisp knowledge
2025, vol.17 , no.2, pp. 65-76
Article [2025-02-07]
Many practical applications, such as medical diagnosis, business decision-making, information searching and retrieval, etc., require usage of uncertain or ambiguous knowledge. Classical ontology-based technologies can represent and reasoning only with cri Several fuzzy or probabilistic extensions of classical description logics and languages for semantic knowledge representation have been proposed recently, but high reasoning complexity of its decision procedures make difficult its usage in real applications. It is of great importance to select the knowledge representation technology, ensuring both the needed expressiveness and effective reasoning. In this paper we make short analysis of knowledge representation and reasoning capabilities of description logics and ontology representation languages allowing representation of uncertain knowledge and propose a methodology for selecting the best description logic and ontology representation variant for every practical application.
Tatyana Ivanova, Petya Petkova. Semantic knowledge models of non-crisp knowledge. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 65-76.
Formulation of a control problem on a discrete uneven scale under conditions of irregular stochastic external influences
2025, vol.17 , no.2, pp. 57-64
Article [2025-02-06]
The article presents an approach to the study of the control problem on a discrete uneven time scale under conditions of irregular stochastic external influences. The purpose is a formalized description of the control process of the managed system, which takes into account the monitoring results and provides the opportunity to find optimal mechanisms for influencing the control object. By choosing certain control mechanisms at a particular moment in time, the subject of control expects that the system, under their influence, will enter a certain state at the next moment in time. In addition, it is assumed that the control object periodically receives external influences, the impact of which is extremely difficult to predict. To form a control action, it is necessary to obtain information about the system states. For this purpose, a monitoring system is provided, which at the necessary time takes results about the conditions of the control object. If the decision maker has the opportunity to spend money on all the mechanisms he plans, then the task will be reduced only to finding the minimum required amount of costs for each of these mechanisms. Otherwise, it is necessary to allocate the available funds in such a way that the effectiveness of choosing all the mechanisms would be maximized. As a result, a multi-criteria optimization problem has been formulated, the solution of which will make it possible to form the best control.
S. A. Oleinikova, A. S. Koptelova, O. Yu. Zaslavskaya, Yu. V. Redkin, P. A. Rahman, S. A. Sorokin, O. Ja. Kravets. Formulation of a control problem on a discrete uneven scale under conditions of irregular stochastic external influences. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 57-64.
Conceptual architecture for imparting AI with olfaction: Integration of electronic noses, aroma generators and LLM
2025, vol.17 , no.2, pp. 45-56
Article [2025-02-05]
This paper proposes a conceptual architecture to integrate olfaction into AI using an electronic nose, an odor generator, and large language models like ChatGPT. The electronic nose detects odors via chemical sensors, while the aroma generator reproduces similar scents remotely. Large language models enhance real-time system calibration and provide a natural language interface, addressing issues like signal standardization, sensor calibration, and data interpretation. Potential applications span industry, medicine, VR/AR, and e-commerce. The outlined workflow and challenges pave the way for prototyping and real-world testing to redefine AI's interaction with the physical and digital worlds.
Aldeniz Rashidov, Fatme Rashidova. Conceptual architecture for imparting AI with olfaction: Integration of electronic noses, aroma generators and LLM. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 45-56.
Optimal control of multifunctional reconfigurable robotic systems
2025, vol.17 , no.2, pp. 33-44
Article [2025-02-04]
This article discusses the optimal control of a robotic system composed of multi-modular robots. Functionally, these robots can perform various tasks using different combinations of modules, making both the robotic system and its individual robots multifunctional. The article examines the operation of a robotic complex in which an individual robot may experience either complete or partial failure. To enhance the efficiency of the robotic complex, we have developed innovative models and algorithms for the optimal selection, distribution of functions, replacement, and interchangeability of robots. In the event of a complete failure, the failed robot is optimally replaced by a backup robot. In the case of a partial failure, optimal robot interchangeability is implemented. To study and evaluate reliability criteria, the article employs optimization methods, logical-probabilistic modelling, and dynamic programming.
Sergo Tsiramua , Sulkhan Sulkhanishvili , Roman Chertovskih , Irakli Basheleishvili . Optimal control of multifunctional reconfigurable robotic systems. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 33-44.
Application of the growing neural GAS algorithm to optimize computational resources in neural network architecture
2025, vol.17 , no.2, pp. 25-32
Article [2025-02-03]
The paper is devoted to the study of methods for improving the efficiency of using and training neural networks, given their complex architectures and a large number of parameters. The author has developed a method for optimizing the number of trained parameters using the growing neural gas algorithm. Approbation results show that the proposed method, combined with convolutional layers, shows a degradation in accuracy compared to traditional convolutional networks, but still outperforms random guessing. Changing the criterion for adding new nodes based on the current error reduced the number of neurons by half. The conclusions indicate that the developed method does not compete with existing solutions, but its further improvement may lead to optimization of computational resources in neural networks.
N. V. Eliseeva, V. E. Petrov, Yu. V. Kaykova, I. V. Kaykova. Application of the growing neural GAS algorithm to optimize computational resources in neural network architecture. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 25-32.
A comparative study of SMOTE and ADASYN for multiclass classification of IoT anomalies
2025, vol.17 , no.2, pp. 15-24
Article [2025-02-02]
The rise of IoT technologies has increased the need for robust threat detection to address growing cyber-physical risks. Traditional machine learning (ML) models often struggle with imbalanced datasets, particularly in detecting rare threats. This study tackles this challenge using the "IoT_Modbus" dataset, a benchmark for multiclass cybersecurity classification, and explores advanced resampling techniques to enhance model performance. By combining methods like Synthetic Minority Oversampling Technique (SMOTE) and Adaptive Synthetic Sampling (ADASYN) with ensemble classifiers, Random Forest (RF), Gradient Boosting, XGBoost, LightGBM, and CatBoost, significant improvements in accuracy were achieved. RF outperformed others, reaching 99.06% accuracy with SMOTE and 99.00% with ADASYN. Among boosting-based models, XGBoost led with 86.21% (SMOTE) and 86.63% (ADASYN), while CatBoost followed closely. These results highlight the effectiveness of resampling in addressing class imbalance and improving predictive performance. Visual tools like heatmaps and accuracy charts further clarify these trends. This research provides cybersecurity professionals with a practical framework to enhance IoT threat detection and underscores the importance of resampling strategies for better classification outcomes. By guiding model selection for imbalanced datasets, it offers valuable insights for securing IoT systems in an interconnected world.
Fayez Alharbi. A comparative study of SMOTE and ADASYN for multiclass classification of IoT anomalies. International Journal on Information Technologies and Security, vol.17 , no.2, 2025, pp. 15-24.