XLIV Simpósio Brasileiro de Telecomunicações e Processamento de Sinais

Underwater acoustic recognition faces challenges in robustness and generalization. This paper provides a unified framework evaluating joint motor--state, motor-only, and state-only classification on the Wolfset dataset. We compare handcrafted features with classical and deep classifiers, analyzing closed-set and open-set (OOD) performance. A specialized...

Point cloud processing is essential for modern robotics and three-dimensional mapping systems. This study investigates the computational efficiency of the Open3D, PyntCloud, and Trimesh libraries when handling point cloud data in PCD, PLY, and LAS formats. The approach evaluates usability, visualization, and algorithmic support under different...

This paper presents an engagement detection system for smart classrooms leveraging a private 5G network and Multiaccess Edge Computing (MEC). The architecture integrates IoT and computer vision to monitor student behavior through four strategically positioned cameras. By offloading video streams to the edge, the system executes classification of student...

Automatic fruit and vegetable recognition is an essential component of intelligent weighing systems in supermarkets and self-service stores. These systems must operate under strict latency and hardware constraints while maintaining high classification accuracy under varying lighting, backgrounds, and occlusions. This work investigates compact...

This article presents a comprehensive approach to forest fire detection, starting with the construction of a deep learning model based on the MobileNet architecture and transfer learning. Post-training quantization is applied to make the model suitable for edge device implementation. It provides computational efficiency, reducing storage size by 64.75%,...

This paper analyzes the statistical behavior of the cyclic prefix (CP)-based artificial noise (AN) from the eavesdropper's perspective to evaluate physical-layer security in power line communication (PLC) systems. The received AN is modeled as a random vector, and its analysis is based on the derivation of first- and second-order statistical moments....

Manual inspection of domestic water tanks is inefficient, often missing critical drops in water availability. We present an end-to-end IoT stack that couples an HC-SR04 ultrasonic sensor with an ESP32 controller, batches 45 samples (at 1 Hz) every 45 seconds, and authenticates payload delivery through a Supabase Edge Function. A React web dashboard...

Este artigo apresenta uma abordagem de localização LoRaWAN baseada em RSSI utilizando um gateway móvel para emular múltiplas âncoras em regiões sem infraestrutura. Experimentos foram realizados em três cenários urbanos reais. O método combina modelo Log-Normal e multilateração por mínimos quadrados. Foram avaliadas técnicas de pré-processamento do RSSI,...

This article investigates the joint optimization of \textit{Power Beacon} (PB) placement and energy \textit{beamforming} in Wireless Energy Transfer (WET) systems for IoT networks, aiming to minimize the total charging time and the devices' \textit{outage} probability. A scenario with multi-antenna PBs, analog \textit{beamforming}, statistical channel...

Este trabalho propõe o uso de códigos BCH como filtro algébrico de validação pós-alinhamento para detecção de mutações pontuais (SNPs) na proteína Spike do SARS-CoV-2. As sequências provêm do banco GISAID, onde são alinhadas contra a referência pelo MAFFT, um método heurístico baseado em FFT que não garante alinhamento exato como algoritmos de...