XLIV Simpósio Brasileiro de Telecomunicações e Processamento de Sinais

Transformer-based neural video codecs achieve strong rate-distortion performance, but their high complexity and reliance on floating-point arithmetic hinder practical deployment. In this work, we propose a module-wise post-training quantization (PTQ) pipeline for the MaskCRT codec, where each module is wrapped to interface with a quantization framework,...

This paper presents a real-time algorithm validation platform on a Raspberry Pi 5 with Bluetooth audio streaming. It supports unidirectional transport through the Advanced Audio Distribution Profile (A2DP) and bidirectional transport through the Hands-Free Profile (HFP), forming a testbed for evaluating algorithms under realistic wireless conditions. The...

This paper reports an FPGA implementation of a quantized neural equalizer for a real-valued MIMO/BPSK link. A synthetic channel emulator generates fixed-channel 2 × 2 and 4 × 4 MIMO data, and a fully connected quantized network is trained with Brevitas, compiled with FINN, and validated on a Xilinx ZCU104 MPSoC. We evaluate input, weight, acti- vation...

Traditional Data Loss Prevention (DLP) tools cannot prevent a smartphone from photographing sensitive information displayed on a monitor. This paper proposes a computer vision-based system that detects smartphones positioned toward a screen and triggers immediate screen obfuscation. Using a standard webcam and a YOLOv8n pipeline implemented in Python on...

This paper investigates parameter-efficient temporal neural architectures for behavioral modeling of RF power amplifiers using physically meaningful feature representations. The investigated architectures were inspired by recent parameter-efficient temporal convolutional network digital predistortion (TCN-DPD) structures proposed in the literature. For...

This article proposes the development of a low-cost assistive device capable of capturing, analyzing, and providing auditory scene descriptions to help users understand and navigate their environment. The goal is to address the gap between the demand for assistive technologies and the availability of accessible devices in Brazil. The research follows an...

Arithmetic Reconciliation (AR) is a recently proposed reconciliation protocol for Continuous-Variable Quantum Key Distribution (CV-QKD) systems operating in low-SNR regimes. Since the channel signal-to-noise ratio (SNR) directly affects the statistical characterization of the induced binary subchannels, inaccuracies in SNR estimation may significantly...

O compartilhamento de espectro licenciado pode aumentar a eficiência de uso do espectro, mas prejudicar usuários primários. Assim, é necessário resolver um problema de otimização que ofereça configurações de equipamentos secundários maximizando a eficiência do espectro sem causar interferências indesejadas a usuários primários. Este trabalho propõe uma...

This paper investigates the feasibility of predicting LTE/5G signal-quality metrics (RSRP, RSRQ, SINR, and RSSI) from purely spatio-temporal inputs (latitude, longitude, hour, weekday) and using the predicted values to recover the ITU/3GPP composite quality label. A drive-test dataset of 17,646 measurements collected in Manaus, Brazil (April 13-16, 2026)...

This paper proposes a tensor-based channel estimation framework for an uplink MIMO system assisted by a movable intelligent surface. The considered architecture combines a fixed transmissive metasurface with a smaller movable layer, whose discrete positions create an additional structured training dimension. By jointly exploiting fixed-layer phase...