Projetos de Pesquisa
Os projetos listados são desenvolvidos pelos professores, doutorandos, mestrandos e alunos do curso de Engenharia Elétrica.
Redes Definidas por Software – RDS (Software Defined Networks - SDN)
Transmissão de Vídeo sobre LTE (4G)
Estratégias de Transmissão para Sistemas de Video sob Demanda
Sistemas de Rádio Cognitivo
Analysis of use of LTE for Smartgrid Communication (closed)
Technology 3GPP (Long Term Evolution, LTE) is considered a promising solution for integration Smartgrid systems because it provides bandwidth and low latency. However, the LTE was not originally designed for applications of this type, which have restrictions on delay and loss. The goal of this project is the analysis of strategies for the use of LTE for Smartgrid communication, including particular issues of implementation of LTE system in Brazil. The specific objectives are: (1) evaluation of performance of Smartgrid using the LTE, (2) evaluation of mechanisms of quality of service of LTE to support Smartgrid applications, (3) evaluation of the effects of background traffic in QoS parameters and (4) evaluation of the LTE uplink scheduling algorithms to support the requirements of the Smartgrid. This part of a large project between many research intitute in Brasil, including Parana Federal University, Parana Technological Federal University, Pontifical University of Parana, Technische Universitat Dresden (Germany), Kungliga Tekniska hogskolan (Sweden). Waiting approval of Brazilian National Research Council. Reseachers: Carlos Marcelo Pedroso (project coordinator), Eduardo Parente Ribeiro, Luis Lolis, Evélio M. G.Fernández, and many others.
Post Date: 2-12-2013
Qualidade de Experiência em Sistemas IPTV (Encerrado)
Monitoring and Evaluation of Projects and Services for 2014 World Cup in Curitiba - Paraná (Technological Development Project) (Closed)
The project has the overall objective to evaluate the technological developments and monitoring activities required for the realization of the FIFA 2014 World Cup in Curitiba. In particular the proposal comes to projects and infrastructure, regarding the physical execution of projects as the stadium of the games of the World Cup (Arena da Baixada) and infrastructure in its surroundings related to the mega event. In this perspective, the UFPR have many teams analysing several themes: the Stadium of Club Atlético Paranaense, Afonso Pena International Airport in São José dos Pinhais, telecommunications, energy, urban mobility, safety, health, and tourism. Researchers: Sergio Sheer (project coordinator), Carlos Marcelo Pedroso (coordinator of subprojects telecommunication and energy), Barbara Mendes, and many many others. Project funded by the Brazilian National Research Council (CNPq).
Post Date: 2-12-2013
Development of a new model for VoIP traffic characterization (Closed)
The classical models used for telephone traffic modeling can not be used for modeling of Voice over Internet communication (VoIP), for two main reasons: (1) the session time in VoIP systems present higher probability of occurrence of extreme values, which differs from the classical approach that uses Markovian models and (2) the traffic profile generated depends heavily on the encoding / decoding (codec) in use. Thus, if the classical theory modeling of voice systems is applied to VoIP, the system tends to underestimate the traffic demand. This project aims to analyze samples collected in real traffic telecom operators and propose a model to characterize a system appropriately VoIP. Reseachers: Carlos Marcelo Pedroso (project coordinator), Eduardo Parente Ribeiro, Carlos Ignacio Mattos, Rafael Alessi Muntsch, Willian Daniel de Mattos, Edgard Masshiro Munetiko.
Post Date: 2-12-2013
Development of a new model for VoIP traffic characterization (Closed)
The IPTV system uses coding algorithms that produce a variable bit rate traffic profile. MPEG4 and MPEG2 exploit temporal redundancy to reduce the amount of information to represent the video.
Although highly desirable, video compression produces a highly variable traffic profile which can cause packet discard in the queues of the network elements.
In this project we show that one can use Artificial Neural Networks to select a packet to discard in order to reduce the quality degradation in case of queue congestion.
Reseachers: Carlos Marcelo Pedroso (Coordinator), Eduardo Parente Ribeiro, Carlos Eduardo Santos Maffini, Felipe Silveira Greco.
Post Date: 2-12-2013