Radioelectronics and Communications Systems

Comprehensive analysis of 5G hybrid FSO communication networks for improved backhaul connectivity (review)

2024-11-18T14:27:59+02:00

Increased capacity, faster data rates, lower latency, and improved service quality are the primary goals in the fifth-generation (5G) and beyond era. Cellular network architectures must be significantly improved to meet these needs. To accomplish these goals, the presented analysis emphasizes the critical roles of 5G cellular network architecture, the substantial use of multiple-input multiple-output (MIMO) technology, and the integration of device-to-device communication. Despite being a promising option, free-space optics (FSO) struggles when used alone in challenging weather conditions. Because FSO systems rely on the line-of-sight connectivity, flying birds and trees do not affect them. Furthermore, unfavorable weather conditions, such as fog and snow, significantly affect FSO performance, resulting in a substantial decline. This study focuses on hybrid FSO/radio frequency (RF) communication systems and provides a comprehensive review of ongoing efforts in this field. Every method used in these models to attain the best possible data throughput and bit error rate (BER) performance for incorporation into 5G networks has been recorded and examined.

Electrodynamic analysis of double-sided microstrip lines for broadband exponentially tapered transitions

2026-09-02T23:29:55+03:00

A method for calculating the dispersion characteristics of double-sided microstrip and balanced transmission lines has been developed. The method is based on a rigorous solution of an internal boundary value problem for the Helmholtz equation in a rectangular domain with a three-layer dielectric filling and two strip lines located on both sides of the dielectric substrate. The results of numerical calculations of the deceleration coefficients and wave impedance of microstrips for typical substrate material parameters (RT/Duroid 5880 and FR4) are presented. As an example, the scattering characteristics of smooth tapered transitions with an exponential profile between microstrip and balanced transmission lines, and the transition between two microstrip lines located on opposite sides of the substrate, were simulated, calculated, and experimentally measured. According to the measurement results, smooth transitions with an exponential profile between transmission lines of different types, calculated using the exact values of the deceleration coefficient and wave impedance for double-sided microstrip lines, enable ultrawideband signal transmission. Experimental results have shown that smooth transitions with an exponential profile in the 1–6 GHz frequency band achieve a reflection coefficient of –15 dB or better.

Identifying LoRa parameters using convolutional neural networks

2026-09-02T23:36:05+03:00

This paper proposes an enhancement to the existing methodology for employing Convolutional Neural Networks (CNNs) to identify parameters of the Long Range (LoRa) telecommunication standard. Building upon the approach of classifying signals using predefined combinations of modulation parameters, our method expands the detectable set by at least twofold through signal-processing modifications. Using the YOLO11 object detection model, our results demonstrate a notable improvement in Mean Average Precision (mAP50) to 93.5%. This surpasses the previously established state-of-the-art result by 2.6%. Notably, our model reveals enhanced detection capabilities on small network sizes. It successfully identifies LoRa signals within spectrograms as small as 256×256, achieving an mAP50 of 87.1% without GPU acceleration, surpassing the 55.0% of the previous research. We validated our technique through an experiment conducted with a First-Person View (FPV) drone. With the drone control signal attenuated to the loss-of-connection level, our model still exhibited remarkable precision. This re-examines the training methodology, demonstrating that a model trained on artificial data can effectively discern modulation parameters from authentically captured signals. The outcomes of our research can be further integrated into a real-time radio intelligence module within discovery and defense systems.

Parametric weighted least squares algorithm for multilateration-based localization

2026-09-02T23:44:14+03:00

Ultra-wideband (UWB) Time-Difference-of-Arrival (TDoA) is a popular approach for indoor system localization due to its simplicity, efficiency, high performance, and low energy consumption. To solve the nonlinear multilateration problem associated with determining the transmitter’s coordinates based on data from synchronized receivers, the least-squares method with low computational complexity or its improved variants, in particular, the two-stage weighted least squares (WLS) method, are often used. However, these methods perform poorly on ill-conditioned matrices due to a specific mutual arrangement between the receiver and the transmitter. This paper proposes a parametric weighted least squares algorithm that is effective under these conditions and enables the determination of an entire family of solutions. It provides accurate and reliable position estimates even under complex conditions associated with system rank reduction. The efficiency of the proposed algorithm is analyzed under various scenarios, including Gaussian noise, synchronization errors at the reference node, and multipath interference, for different receiver configurations arranged in a square grid.