For access to this article, please select a purchase option:
Buy conference paper PDF
Buy Knowledge Pack
IET members benefit from discounts to all IET publications and free access to E&T Magazine. If you are an IET member, log in to your account and the discounts will automatically be applied.
Since the publication of Krizhevsky et. al. paper on image classification using convolutional neural networks, and the progress in so-called deep learning that followed, researchers are exploring applications in non-artificial intelligence areas. This work describes the application of a three-layer feedforward neural network for classifying wireless communication scenarios derived from a geometric stochastic channel model. Simulation results show the network's ability to classify four scenarios having distinct ensemble average power delay profiles with successful prediction of 90% on the test set. If the scenario conditions are modified so that the ensemble average power delay profiles are no longer distinct, the model accuracy on the test set drops to 54%. Thus, the viability of the classification depends on the information (features) that can be extracted from the data input to the neural network.