System research and information technologies

Implementation of a generalized intermittency scenario in the Rössler dynamical system

2026-04-19T23:13:40+03:00

The realization of novel scenario involving transitions between different types of chaotic attractors is investigated for the Rössler system. Characteristic features indicative of the presence of generalized intermittency scenario in this system are identified. The properties of “chaos–chaos” transitions following the generalized intermittency scenario are analyzed in detail based on phase-parametric characteristics, Lyapunov characteristic exponents, phase portraits, and Poincaré sections.

Multimodal system for skin cancer detection

2026-04-19T21:45:37+03:00

Melanoma detection is vital for early diagnosis and effective treatment. While deep learning models on dermoscopic images have shown promise, they require specialized equipment, limiting their use in broader clinical settings. This study introduces a multi-modal melanoma detection system using conventional photo images, making it more accessible and versatile. Our system integrates image data with tabular metadata, such as patient demographics and lesion characteristics, to improve detection accuracy. It employs a multi-modal neural network combining image and metadata processing and supports a two-step model for cases with or without metadata. A three-stage pipeline further refines predictions by boosting algorithms and enhancing performance. To address the challenges of a highly imbalanced dataset, specific techniques were implemented to ensure robust training. An ablation study evaluated recent vision architectures, boosting algorithms, and loss functions, achieving a peak Partial ROC AUC of 0.18068 (0.2 maximum) and top-15 retrieval sensitivity of 0.78371. Results demonstrate that integrating photo images with metadata in a structured, multi-stage pipeline yields significant performance improvements. This system advances melanoma detection by providing a scalable, equipment-independent solution suitable for diverse healthcare environments, bridging the gap between specialized and general clinical practices.

The principles of synthesizing the coordinating automatic control systems

2026-04-15T23:31:07+03:00

The problem linked to the model development of the coordinating automatic control system has been solved in this scientific work. Such a task is relevant, as it is linked to increase the operational efficiency of the technological object due to the quality improvement of ratio control. As an object the steam boiler has been considered, in which it is necessary to maintain the appropriate air-fuel ratio. The article describes the stages of the model development for the coordinating automatic control system ensuring the maintenance of a given ratio between controlled variables in transient modes. The support of the set ratio is ensured by the implementation of temporal subordination for control processes. Based on the analysis of modeling results, we have made the conclusion about the expediency of using the coordinating system of automatic control. Thus, the method of stepwise synthesis for coordinating control systems was further developed.

Improving the SOM algorithm to ensure stability and reproducibility of data clustering results

2026-04-19T23:42:13+03:00

The article proposes a method to improve the Kohonen Self-Organizing Map (SOM) learning algorithm to ensure the stability and reproducibility of clustering results, an urgent task when working with large amounts of data. SOM is widely used in clustering and visualization tasks, especially in applications that require analyzing multidimensional data structures, such as telecommunications billing systems and financial analysis. The standard SOM implementation, which includes random weight initialization and stochastic sample selection during training, leads to significant cluster variability even when using the same input data and identical network training parameters. This makes it difficult to apply this algorithm in cases where stability and reproducibility of results are required. To solve this problem, we propose modifying the algorithm to include its own random number generator and introducing a seed parameter to fix the initial training conditions. This reduces variability and ensures reproducible clustering results, thereby increasing the reliability of the analysis and the suitability of the SOM algorithm for real business tasks. The proposed method has been tested on data from billing systems, where the reproducibility of clustering results is critical for effective work with customer segments, the development of targeted marketing strategies, and the creation of personalized tariff plans.

Long-term monitoring of surface water quality and groundwater po-tential using computational intelligence, GIS technologies, and remote sensing

2026-04-19T23:53:19+03:00

Water scarcity and declining water quality due to population growth, urbanization, industrialization, and climate change highlight the importance of effective water management. Advances in remote sensing, cloud computing, and computational intelligence underscore the need to utilize modern technologies for monitoring surface water quality. This research involves the development of hybrid intelligent models using Landsat and Sentinel-2 images and WISE data with hybrid deep learning networks to evaluate surface water quality and groundwater potential. Correlation analysis revealed strong connections between remote sensing data and water quality parameters (such as chlorophyll-a, dissolved oxygen, nitrogen, and phosphorus). The hybrid models surpassed traditional machine learning methods, demonstrating their effectiveness in real-world water management.

Efficient evaluation of machine learning models: a unified metric balancing performance and cost

2026-04-20T00:08:03+03:00

This paper introduces a novel, unified metric for evaluating the efficiency of machine learning, deep learning, and artificial intelligence models by balancing predictive performance and execution cost. Existing metrics typically isolate performance or execution measures (e.g., FLOPs, latency, energy), failing to capture the inherent trade-off between resource constraints and predictive capability in single formula. The proposed formula incorporates a tunable trade-off factor and hard constraints on performance and cost, allowing principled comparison across models and deployment settings. Our formulation generalizes prior heuristics and demonstrates clear interpretability, scalability, and hardware awareness.

Physical-informed neural network in signal processing and network traffic communications

2026-04-20T00:24:56+03:00

Problem of signal processing and network traffic optimization is solved at the hardware level and is interesting, modern and relevant from the point of view of the application level. It is necessary to propose an approach that combines machine learning methods with network bandwidth tasks and traffic over the network. To solve this problem, it is proposed to use concept of Informed Machine Learning (IML), that is the Taxonomy of IML, the principles of constructing deep machine learning systems based on information about the physical properties of the data transmission network under study. The platform for developing is deep machine learning models using PINN neural networks. The PINN represents the class of deep learning algorithms that can integrate data with or without physical processes description. As an algorithm it is proposed to use the popular algorithm in the field of deep learning – ADAM (Adaptive Moment Estimation) for optimizing network traffic. Using the PINN trained with the ADAM algorithm to transmit data the efficiency has increased. Thanks to this method, it is possible to obtain a low noise signal in practice, due to which network traffic is optimized.

A probabilistic model of the Colonel Blotto game without symmetry and homogeneity constraints

2026-04-19T22:58:22+03:00

The classic Colonel Blotto game for two players was considered. The probabilistic model of the payoff functions of the specified problem was investigated, and the game conditions are not subject to the restrictions of symmetry and homogeneity. The system of equations obtained using the method of Lagrange multipliers has a large dimension. In order to find a solution, a way to reduce the dimension was found. The found ratio between the resources of both players, distributed over the courts, made it possible to identify a parameter determined by the ratio of Lagrange multipliers from the corresponding functions for both players. For such a parameter, an interval constraint that it satisfies was found, and an equation is formulated to find it, which is solved numerically. The found value of the parameter makes it possible to calculate individual Lagrange multipliers and obtain the optimal distribution of players’ resources in the form of a Nash equilibrium in pure game strategies. An example of a game under significantly different conditions for players was studied.

Polynomial-based method for linearizing the temperature response of NTC thermistors

2026-04-19T22:05:59+03:00

The objective of this research is to develop a circuit-based method for linearizing the temperature characteristics of thermistors using a polynomial digital technique for NTC-thermistor temperature characteristics, together with the characteristics of the designed measurement channel, by means of an original MATLAB Simulink model. A model of a temperature-measurement device employing a thermistor-based sensor is proposed. The polynomial digital method for linearizing the temperature characteristics of NTC thermistors in the developed device has been simulated. To improve measurement accuracy, the components of the measurement channel were selected such that they enable simulation with minimal error. Methods for introducing correction-coefficient values into the model have been determined to compensate for error-inducing factors, including the thermistor’s self-heating effect. The proposed data-processing algorithm offers advantages for implementation in a low-cost, low-power microcontroller.

Practical aspects of creating a data transmission system for controlling unmanned surface vehicles in unstable communication channels

2026-04-19T22:27:58+03:00

The study presents the development and verification of an adaptive data transmission system for controlling unmanned surface vehicles (USVs) in unstable communication channels. The work aims to overcome the limitations of existing technologies, which include LTE networks and satellite systems that fail to deliver stable service quality for USV remote control operations. The proposed adaptive routing algorithm evaluates communication channel status through three vital indicators, which include delay, packet loss, and availability. The algorithm selects the best channels according to changing weight parameters. Experimental results confirmed a significant reduction in data transmission delays, stable real-time video streaming with a delay of 1–4 seconds, and a reduction in packet loss to below 2 %. In addition, the system implements the use of modern video coding standards (e.g., H.265) and secure VPN channels, which increase bandwidth efficiency and the level of cybersecurity. The results confirm the practical suitability of the proposed system for USV operation in real marine conditions, as well as its potential for use in critical scenarios that require stable, low-latency communication.

Optimizing microservices design pattern: maximizing communication speed and prolonging application longevity

2026-04-19T20:28:08+03:00

Microservice oriented application design obtained popularity in the past years. Most researchers investigated some aspects in microservice design for implementing application functionality. Little research considered the core functionality of microservices. This research investigates how to construct a microservice communication system by yourself in detail. The results should assist developers and architects to construct their own microservice applications, use less amount of frameworks and therefore prolong overall microservice system life cycle. The standard TCP/IP connection and embedded libraries were used to construct the communication system without using any additional frameworks. As a practical application of this methodology a microservice core system was implemented with a minimum number of microservices to perform performance testing. The measured application layer communication speed turned out to exceed the speed in real application because of database operation limitations. The implemented microservice core system is intended to be used in financial commercial applications as well as in further scientific investigations.