Robert Corrigan’s scientific contributions

... O segundo motivo para a utilização das redes neurais é a grande disponibilidade de ferramentas que, em virtude do aumento da capacidade computacional disponível atualmente, conseguiram viabilizar o treinamento e utilização de redes neurais com resultados bastante satisfatórios para as aplicações de comunicação [19]. ...

Reference:

Utilização de Técnicas de Aprendizado de Máquina para Demodulação de Sinais

... As far as artificial neural networks are concerned in this paper, neurons are usually modeled as non-linear functions between channel characteristics and scenarios, which employ a plurality of realvalue inputs representing received signals coming from neighboring neurons and calculate the strength of the output signals of the neurons. The outputs are calculated by taking weighting sums of all inputs and applying an activation function to the result [53]. The weights themselves finally endow the artificial neural network with the ability of scenario identification, because these weights can be adjusted to output corresponding scenarios according to the input channel characteristics. ...

Reference:

Machine-Learning-Based Scenario Identification Using Channel Characteristics in Intelligent Vehicular Communications