Cesar Vargas-Rosales of Tecnologico de Monterrey

In this conference, we compare two artificial neural networks (ANNs) approaches designed to perform channel equalization for millimeter-wave (mmWave) signals operating in the 28 GHz frequency band. We used an in-house deterministic Three-Dimensional Ray-Launching (3D-RL) code to simulate the spatial structure of mmWave channels considering the material properties of the obstacles within the scenario at the frequency under analysis. We performed offline training of a multilayer perceptron (MLP) neural network with the simulated mmWave channels to equalize the received signal. We also performed online training of an extreme learning machine (ELM) neural network to directly get the equalized symbols at the receiver, given as input the received mmWave signal. The ANN solutions were tested in terms of the achievable spectral efficiency, bit error rate, and time to process. We compared the ANN techniques to the minimum mean square error and the zero-forcing equalizers, considering an orthogonal frequency-division multiplexing communication based on the 5G New Radio standard. We present numerical results on the performance of the proposed ANNs and show that the ELM strategy outperforms the MLP method, requiring significantly less processing time than the reviewed equalization methods.

Biography:

Cesar Vargas-Rosales is currently a professor at Tecnologico de Monterrey, in Monterrey Mexico, and the Leader of the Strategic Research Group on Telecommunications. He received the Ph.D. and M.Sc. degrees in Electrical Engineering with a major in Communications and Signal Processing, from the Department of Electrical and Computer Engineering at Louisiana State University, Baton Rouge, LA.

Dr. Vargas-Rosales is a member of the Mexican National Researchers System (SNI level II), the Academy of Engineering of Mexico (AIM) and the Mexican Academy of Sciences (AMC). He has been the IEEE Communications Society Monterrey Chapter Chair and the Faculty Advisor of the IEEE-HKN Lambda-Rho Chapter. He has also been the Technical Program Chair of the IEEE WCNC. He serves on several editorial boards and program committees of prestigious IEEE journals and conferences. His research interests are personal communications, 5G, cognitive radio, MIMO systems, stochastic modeling, intrusion/anomaly detection in networks, position location, interference, error correcting codes and reconfigurable networks.

Email:

Address:Monterrey, Mexico

Vinicio Carrera of Universidad de las Fuerzas Armadas ESPE, Matriz

Esta charla describe las principales características que diferencian al aprendizaje no-supervisado del supervisado y las nuevas aplicaciones que ha generado el desarrollo de algoritmos no-supervisados sustentados principalmente en lo que que se denomina aprendizaje profundo.  Se detallan básicamente los denominados autoencoders y las redes adversarias generativas, así como sus usos más comunes hoy en día.

Biography:

Es Ingeniero electrónico en la ESPE, tiene maestría en ingeniería eléctrica en la Universidad Católica de Rio de Janeiro, y es Doctor en Ciencias en Ingeniería de Computación en la Universidad Federal de Rio de Janeiro. Fue investigador invitado en la Universidad de Rochester, EEUU, y también investigador asociado en la Universidad de Rutgers, EEUU.  Trabajó como profesor asociado en la USFQ, y ahora es profesor titular principal en la ESPE.  A partir de 2014 ha colaborado como profesor externo en la Universidad Rey Juan Carlos, España, y es miembro Senior de IEEE, así como Visitante Distinguido por la Sociedad de Computación del IEEE.  Tiene alrededor de 100 publicaciones en reconocidas revistas y conferencias. Sus principales áreas de investigación son el procesamiento digital de señales y la inteligencia computacional.

Email:

Address:Quito, Ecuador