Radioelectronics and Communications Systems

Features of asymptotic method implementation for electromagnetic scattering problem in material with chaotically distributed inclusions

2026-04-11T22:33:28+03:00

The electromagnetic (EM) scattering problem on a set of randomly distributed particles in a homogeneous material is solved by an analytical-numerical asymptotic method. This is possible due to the assumption of small particles. The analytical part of the solution consists of deriving an approximate formula for a system of linear algebraic equations (SLAE) with respect to the values of an unknown auxiliary function. The numerical solution of the auxiliary SLAE allows us to solve the initial scattering problem and to derive an explicit formula for the magnetic permeability of the resulting inhomogeneous material. The software implementation features of the proposed method are described in detail, including heterogeneous region geometry simulation, complexity, and computation time for individual key formulas. Optimal method selection for direct problem solving in terms of computational resources is an essential prerequisite for the successful solution of the inverse problem, since heuristic algorithms require multiple solutions to direct problems with different input parameters. Numerical data indicate the possibility of obtaining a more diverse distribution of magnetic permeability in the resulting heterogeneous material compared to regular and chaotic methods of particle embedding.

Improving network efficiency with antenna and user selections in 5G heterogeneous cellular network

2024-08-11T19:47:30+03:00

The ever-increasing demand for 5G networks necessitates innovative solutions to optimize network efficiency. In this study process, we approached conventional Random User Selection (RUS) and Maximum Channel Gain (MCG) based on which we developed a new method called Distance User Selection (DUS). In DUS, the users are selected based on the distance where the nearest user is chosen first and assigned to the available antennas at the base station. This method prioritizes users closer to the serving base station for association, exploiting stronger signal strength and reducing path loss. Additionally, it employs optimized antenna selection algorithms to improve the signal quality and resource allocation further. We evaluated the proposed method through simulations and compared its performance with conventional approaches regarding throughput, sumrate, and energy efficiency. Our results demonstrate significant improvements in network efficiency, highlighting the potential of DUS and antenna selection to enhance 5G network performance and user experience.

Classification of low-amplitude ECG components using adaptive activation functions of neural networks

2026-04-11T22:38:32+03:00

A promising direction in the development of neural networks for the analysis and classification of biomedical signals is the use of trainable activation functions, known as AAFs (Adaptive Activation Functions). The use of such functions enables heterogeneous data to be adapted, thereby improving classification accuracy. This paper considers the application of AAFs for the classification of low-amplitude components of electrocardiogram (ECG), specifically ventricular late potentials (VLP) and atrial late potentials (ALP), which are important for the early detection of cardiac tachyarrhythmias. To evaluate the impact of AAF on the quality of VLP and ALP detection, two fully connected neural networks with different numbers of hidden layers were developed. The study established that using AAF increases the accuracy of VLP and ALP classification and the speed of neural network model training compared to non-adaptive activation functions. To minimize the problems of “vanishing” or “exploding” gradients in the loss function, as well as the effects of “dead” neurons that arise during neural network training, a new activation function has been developed that normalizes weight coefficients, preventing excessively high or low gradients. Using the developed activation function increases the speed and stability of neural network training. It improves the recognition accuracy of low-amplitude ECG components compared to other activation functions. Using the developed AAF, the highest classification accuracy was obtained for VLP (94.7%) and ALP (91.4%). To simultaneously analyze a large number of activation functions, a coefficient was developed to assess the redundancy of network layers. The proposed coefficient for detecting “bottlenecks” in neural network architectures significantly simplifies the analysis and improvement of neural networks.

Parameters of ribbon electron beam formed by HVGD guns: study of focal parameters

2026-04-11T22:42:38+03:00

In the second part of the article, the electric field distribution, electron beam trajectories, and their focal parameters are calculated using the known analytical expressions to determine the plasma boundary position relative to the cathode surface. The sequential upper relaxation method, the current tube method, the fourth-order Runge–Kutta method, and extreme analysis numerical methods are used to determine the electric field distribution in the electrode system, to calculate the spatial charge, to calculate the electron trajectories in the free drift region in the anode plasma, and to estimate the thickness of the electron beam in focus, respectively. The iterative calculation relations for the electric field distribution and electron beam trajectories are expressed as arithmetic-logical expressions and recurrence matrices to simplify program code implementation. The obtained relative difference between calculated and experimental data for the thickness of the electron beam in focus doesn’t exceed 15–20%. The theoretical and experimental results obtained are important for further engineering development of electron beam technological equipment intended for industrial use.