Linha de Visada e Alocação de Fator de Espalhamento em Redes LoRaWAN

Lucas Lima Oliveira, Alvaro Augusto Machado de Medeiros, Vicente Sousa, Gabriel Maia de Macedo

DOI: 10.14209/sbrt.2025.1571151696

Keywords:

Abstract

A caracterização do canal sem fio em função da linha de visada tem o potencial de aprimorar a alocação eficiente do fator de espalhamento em redes LoRaWAN. Ao adequar a alocação de fator de espalhamento e potência de transmissão às condições de cada canal, observa-se um aumento na taxa de entrega de pacotes. Baseado em simulações no software ns-3 com dados de campanha de medição, resultados indicam que a identificação de visada permite adoção de estratégias mais simples, porém mais eficientes do ponto de vista energético e de desempenho.

Download

Sobre o Impacto de Eventos Solares no Desempenho da RAN de Sistemas 5G

Samuel Alvim da Silva, Robert Mota Oliveira, Lisandro Lovisolo

DOI: 10.14209/sbrt.2025.1571151697

Keywords: Eventos solares Telecomunicações 5G NR

Abstract

In this article, we analyze the effects of solar radiation on mmWave (25-60 GHz) radio links, both in isolation and when integrated into the 5G mobile network. We found that variations in solar flux deteriorate the signal-to-noise ratio and increase path loss - an impact that tends to worsen in Brazil due to the South Atlantic Magnetic Anomaly. Through data interpolation and simulations, we identified interference patterns that compromise the reliability of the links, and these experiments provided estimates of the degree of degradation within the operating band and the extent to which the 5G network is affected.

Download

Antena de Microfita Dupla Faixa de Baixo Custo Otimizada para as Bandas L1/L5 do GPS

Herick R. da Silva Rodrigues, Marcos V. T. Heckler, Edson R. Schlosser

DOI: 10.14209/sbrt.2025.1571151698

Keywords: Antena de microfita dupla faixa polarização circular GPS

Abstract

O presente artigo apresenta uma antena de microfita quadrada, composta por duas camadas do laminado FR4 e dois patches truncados, projetada para operar com Polarização Circular à Direita (RHCP) nas bandas L1 (centrada em 1,575 GHz) e L5 (centrada em 1,176 GHz) do Sistema de Posicionamento Global (GPS). A antena foi excitada por meio de um único conector coaxial posicionado verticalmente no irradiador, com a adição de um capacitor em série, para obtenção de coeficiente de reflexão inferior a -10 dB em ambas as bandas. Após a modelagem da estrutura no simulador eletromagnético ANSYS HFSS, foi possível otimizar a razão axial, obtendo-se um diagrama de irradiação com elevado ângulo de meia-potência, com elevada supressão de polarização cruzada e alta eficiência de irradiação. Como resultado, a antena apresentou ganho de aproximadamente 7,44 dBic na banda L1 e 5,60 dBic na banda L5. Os resultados detalhados são apresentados ao longo deste artigo.

Download

Deep Learning-Based OSNR Estimation from Constellation Diagrams of DP m-PSK and DP m-QAM in Flexible Coherent Optical Receivers

Mateus F de Araújo, Myke Valadão, Antonio M. C. Pereira, Éderson R. da Silva, Waldir Silva, André L. A. Costa

DOI: 10.14209/sbrt.2025.1571151702

Keywords: OSNR estimation constellation diagrams deep learning coherent optical communications

Abstract

Accurate OSNR estimation is essential for maintaining signal quality in coherent optical communication systems. This work proposes a deep learning-based method for OSNR estimation using constellation diagrams as input. A dataset of over 19,000 images was generated through simulations with various modulation formats. We evaluated 15 CNN architectures, including MobileNetV3, ConvNeXt, DenseNet, and EfficientNet. ConvNeXtBase achieved the best results, with a MAE below 0.43 dB and R2 above 0.98. The results demonstrate the effectiveness of computer vision models for accurate and non-intrusive OSNR prediction.

Download

Detecção Inteligente de Incêndios Florestais Usando Modelos de Aprendizado Profundo

Tarciso Gregório B. de Bello, Samuel Mafra

DOI: 10.14209/sbrt.2025.1571151710

Keywords: YOLOv11 IoT Detecção de Incêndio

Abstract

Este estudo apresenta uma abordagem para detecção inteligente de incêndios florestais utilizando visão computacional e Internet das Coisas (IoT), sendo seu foco no monitoramento da progressão e regressão dos focos de incêndio. A técnica proposta é baseada no algoritmo YOLOv11, que se destaca pela passagem da imagem na rede neural uma única vez, velocidade e precisão na identificação de objetos. A integração com IoT possibilita uma coleta contínua de dados em tempo real, mesmo em áreas de difícil acesso. Foram utilizadas métricas de desempenho como Acurácia, Precisão, Recall e F1-Score para avaliar a eficácia do modelo proposto.

Download

On Performance of Massive MIMO Systems with Solid-state Amplifiers

João Vítor Correia Pessoa, Rafael Chaves

DOI: 10.14209/sbrt.2025.1571151722

Keywords: Massive MIMO analog front-end power amplifiers BER

Abstract

This paper analyzes the impact of analog front-end nonlinearities on the bit-error rate (BER) performance of massive multiple-input multiple-output (MIMO) systems, mainly focusing on the power amplifiers (PAs) nonlinearities of the solid-state am- plifier. Simulation results show that the PA nonlinearities degrade system BER, yielding SNR losses of approximately 2.3 dB to 5.1 dB depending on the number of user equipment, modulation order, and precoding or decoding schemes employed. The findings confirm the necessity of nonlinear compensation to guarantee the viability of massive MIMO in practical applications.

Download

Neural Estimation of Information-Theoretic Generalization Bounds: Limitations and Guidelines

Nathalia Viana, Eduardo N Velloso, Max H. M. Costa, José Cândido Silveira Santos Filho

DOI: 10.14209/sbrt.2025.1571151723

Keywords: mutual information generalization error bounds neural estimation bias-variance decomposition

Abstract

We investigate practical challenges of estimating information-theoretic generalization bounds using neural mutual information estimators. Focusing on a Gaussian mean estimation task, we compare input--output (MI), individual-sample (ISMI), and conditional (CMI) formulations under varying sample sizes and regularization strategies. Through empirical analysis, we identify underfitting regimes, characterize the bias--variance behavior across estimators, and highlight sample complexity ceilings that limit estimation accuracy. Our results provide practical guidelines for selecting estimators and tuning Monte Carlo parameters to achieve reliable generalization bounds in low-data~settings.

Download

Metasurface Absorber for Reduction of Specific Absorption Rate (SAR) at 5G n78 and n258 Communication Frequencies

Mário D. A. Tavares, Roberto B. Di Renna, Maurício Weber

DOI: 10.14209/sbrt.2025.1571151726

Keywords: Specific absorption rate (SAR) absorber metasurface

Abstract

In this study, the reduction of human electromagnetic exposure to millimeter waves is analyzed. Thus, the specific absorption rate (SAR) from mobile antennas used for next-generation 5G n78 and n258 communication frequencies is significantly decreased by an absorber based on periodically arranged metallic square spirals. The reduction of SAR with the optimized metasurface was investigated by CST Microwave Studio in layered human tissues by examining SAR with gram mass averaging. The proposed absorber exhibits an SAR reduction of 99.99% at 3.5 GHz and more than 80% at 26 GHz. The results are aligned with the safety compliance of the 5G user equipment.

Download

Kernel-Based Digital Predistortion: An Approach with the EX-QKRLS Algorithm

Isaac Macario da Silva de Gouveia, José Antonio Apolinário Jr., Claudio Augusto Saunders Filho

DOI: 10.14209/sbrt.2025.1571151729

Keywords: Digital Predistortion RF Weblab Power Amplifier Kernel

Abstract

This paper investigates the effectiveness of kernel-based adaptive filtering for digital predistortion in power amplifiers. Specifically, the EX-QKRLS algorithm, which incorporates concepts from the Kalman filter, is employed to model and compensate for the nonlinearities introduced during signal amplification, without requiring memory polynomials or lookup tables. This method reduces overall signal distortion and improves transmission fidelity robustly and efficiently. Unlike conventional approaches, the proposed technique offers high adaptability, enabling rapid adaptation to changing system conditions, including temperature fluctuations and component aging.

Download

Dual-antenna passive synthetic aperture DOA estimation for airborne platforms

Leandro Geraldo da Costa, Felix Antreich, Daniele Oliveira Silva, Romildo Henrique de Souza

DOI: 10.14209/sbrt.2025.1571151773

Keywords: Passive radar synthetic aperture antenna arrays direction-of-arrival

Abstract

This work presents a dual-antenna passive synthetic aperture direction-of-arrival (DoA) estimation approach for compact airborne sensors. A physics-based model is employed, incorporating spherical wave propagation, Doppler effects, and a 200 km line-of-sight, and is implemented in an in-house radar simulator. The corresponding Cramér-Rao lower bound (CRLB) is derived to analyze the estimator's behavior and, in particular, to explore the trade-off between aperture length Q and number of snapshots K. Computer simulations conducted over a 0-50 dB SNR range demonstrate high azimuth and elevation accuracy, convergence of the root-mean-squared (RMS) error to the theoretical lower bound, and robustness. The proposed scheme offers a lightweight, low size, weight, and power (SWaP) solution for passive airborne sensing platforms.

Download