Repositório referente ao artigo "DTMCash: explorando redundância de viewports de usuários para otimização de streaming de vídeo 360°" publicado no SBRC 2024

User Equipment for non-3GPP access via N3IWF | Working in-progress

This work investigates the implementation of FL algorithms in wireless IoT networks, specifically addressing the challenges of power allocation and communication resource scheduling, considering resource constraints and the occurrence of failures in the training process.

Repositório referente ao artigo "Algoritmo para alocação inteligente de recursos em cluster kubernetes"

FL-E2WS is a Federated Learning algorithm designed as an optimization problem aimed at minimizing the energy consumption of devices in a wireless IoT network while maintaining the quality of the global model.

FL-wDQN é um algoritmo baseado em Deep Q-Network (DQN) para a otimização da alocação dos recursos de comunicação para tarefas de FL em redes IoT sem fio.

Repository describing the experiments associated with the doctoral thesis whose main theme is the superconvergence of non-3GPP wireless communication technologies and the 5G core - pathways to superconvergence.

This work presents FL challenges based on the simulation of a HAR application. As a solution, two data sharing strategies were analyzed in order to reduce the asymmetry of device data distribution.