卷 26, 编号 3 (2025)

The paper discusses the scientific rationale for the structural and parametric synthesis of low Earth orbit communication systems (LEOCSs) with broadband inter-satellite links (ISLs) and on-board data routing, principles and stages underlying data routing on board satellites and the construction of reliable and broadband inter-satellite links. The aim of the study is to specify a structural and parametric synthesis of low-orbital communication systems with broadband inter-satellite links. The object of the study is the constellation of satellites in low-altitude orbits with on-board data routing. The subject of study is the process of synthesis of the structure of the corresponding low-orbital system. The methodology of the study includes mathematical apparatus of queuing systems, used as part of the system analysis toolkit. The scientific significance of the research lies in the development of principles and methods that enable structural and parametric analysis and the development of promising satellite communication systems. Additionally, the research involves the modeling of the behavior of low-orbit communication systems with broadband inter-satellite links and the assessment of their effectiveness. The practical significance of the research is to provide a list of tasks for the creation of domestic satellite constellations that provide personal subscriber communications with data transmission from subscriber to subscriber without intermediate signal landing.

This paper explores various aerial platforms for in-situ atmospheric exploration of Venus, emphasizing their potential integration into future missions. Platforms under consideration include fixed-altitude balloons, variable-altitude balloons, aircraft-like vehicles with three-dimensional maneuvering capabilities, and others. Design configurations of descent vehicles and deployment strategies for these platforms in Venus’ atmosphere are discussed. Specific deployment mechanisms for balloons are detailed. The study also models the dynamics of spherical descent vehicles equipped with balloons, analyzing trajectory parameters during different phases. Results confirm the parameters remain within acceptable limits throughout descent.

This study investigated the feasibility of applying machine learning methods, specifically generative models, for semantic editing of satellite imagery. The research focused on an architecture based on diffusion models capable of generating desirable objects directly on satellite images. However, significant shortcomings were identified in the standard model with regard to realism and relevance to the surrounding context, given the specific nature of the chosen subject area, namely the generation of realistic images of oil and gas infrastructure objects (such as pipelines). To address this limitation, fine-tuning of the neural network was performed. The objective of the fine-tuning was to enhance the quality of visualizing pipeline-related design solutions. A methodological approach for creating training dataset was proposed and described in detail. Based on actual pipeline routes, spatially referenced vector layers were created in QGIS, and a set of satellite image tiles with precise pipeline boundary annotations was generated. The results of the experimental fine-tuning demonstrated a significant improvement in the quality of generated images depicting oil and gas infrastructure objects in satellite imagery compared to the original, non-adapted model. The developed fine-tuned model enables highly realistic pipeline generation, effectively integrating them into the existing landscape within the image. Visual comparison of results before and after fine-tuning confirms the elimination of artifacts and the achievement of the required level of detail. This work demonstrates the effectiveness of the approach involving the creation of specific datasets and fine-tuning for solving specialized visualization tasks in remote sensing.

The construction of a class schedule of an educational institution and, especially, a multilevel higher education institution, combining in its organizational and pedagogical structures several levels of education, including professional, secondary vocational and higher education, as well as training of scientific and pedagogical staff of higher qualification, is a time-consuming task. The study considers a computerized approach to the process of building a model of its optimization. The study uses the methods of system analysis and modification of genetic algorithms (GA), substantiates the structure of initial data for the task of compiling and optimizing training schedules using the method of penalty functions to account for resource and other constraints. A statistical approach is proposed, and a statistics collection and visualization module is implemented, which allows for the operative correction of hyperparameters and the mathematical model of the GA. The examples are provided to illustrate the problem of creating schedules for a multilevel university using GA. The developed computer program provides the creating of the schedule of academic classes of a multilevel university, effective according to the integral quality criterion substantiated taking into account the limitations.

Breast cancer remains one of the leading causes of morbidity and mortality among women worldwide. Despite the global emphasis on early detection, breast cancer continues to pose a significant public health challenge. The object of this study is to predict the breast cancer risk using various machine-learning approaches based on demographic, laboratory, and mammographic data. It employed a quantitative research design to assess the potential of machine learning (ML) in predicting breast cancer. It integrated supervised ML algorithms, including Support Vector Machines (SVM), Decision Trees, Random Forests, and Deep Learning models, to evaluate their accuracy, efficiency, and applicability in medical diagnostics. The dataset revealed significant variability in tumor features such as mean radius, mean texture, mean perimeter, and mean area. The target variable demonstrated a class imbalance, with 62% benign and 38% malignant cases. Among the evaluated models, Random Forest outperformed others with the highest accuracy, precision, recall, F1-score, and ROC-AUC, indicating superior predictive capability. The Logistic Regression and Support Vector Machine models showed competitive performance, particularly in precision and recall, while the Decision Tree model exhibited the lowest overall performance across metrics.