Machine Learning in NextG Networks via Generative Adversarial Networks

Generative Adversarial Networks (GANs) implement Machine Learning (ML) algorithms that have the ability to address competitive resource
allocation problems together with detection and mitigation of anomalous behavior. In this talk, we discuss their use in next-generation NextG) communications within the context of cognitive networks to address:
i) spectrum sharing,
ii) detecting anomalies, and
iii) mitigating security attacks.
GANs have the following advantages. First, they can learn and synthesize field data, which can be costly, time consuming, and nonrepeatable. Second, they enable pre-training classifiers by using semisupervised data. Third, they facilitate increased resolution. Fourth, they enable recovering corrupted bits in the spectrum. The talk will provide basics of GANs, a comparative discussion on different kinds of GANs, performance measures for GANs in computer vision and image processing as well as wireless applications, a number of datasets for wireless applications, performance measures for general classifiers, a survey of the literature on GANs for i)–iii) above, some simulation results, and future research directions. In the spectrum sharing problem, connections to cognitive wireless networks are established. Simulation results show that a particular GAN implementation is better than a convolutional autoencoder for an outlier detection problem in spectrum sensing.
Co-sponsored by: Instituto Politécnico Nacional – Unidad Profesional Interdisciplinaria de Ingeniería "Alejo Peralta"
Speaker(s): Ender Ayanoglu
Bldg: Unidad Profesional Interdisciplinaria de Ingeniería "Alejo Peralta", Instituto Politécnico Nacional, Calle 11 Sur 12122, San Francisco Mayorazgo, Puebla, Puebla, Mexico, 72480