Machine Learning in NextG Networks via Generative Adversarial Networks

Generative Adversarial Networks (GANs) are Machine Learning (ML) algorithms that have the ability to address competitive resource allocation problems together with detection and mitigation of anomalous behavior. In this paper, we investigate their use in next-generation (NextG) communications within context cognitive networks i) spectrum sharing, xmlns:xlink="http://www.w3.org/1999/xlink">ii) detecting anomalies, xmlns:xlink="http://www.w3.org/1999/xlink">iii) mitigating security attacks. GANs following advantages. First, they can learn synthesize field data, which be costly, time consuming, nonrepeatable. Second, enable pre-training classifiers by using semi-supervised data. Third, facilitate increased resolution. Fourth, recovery corrupted bits spectrum. The paper provides basics GANs, a comparative discussion on different kinds performance measures for computer vision image processing as well wireless applications, number datasets general classifiers, survey literature xmlns:xlink="http://www.w3.org/1999/xlink">i)–iii) above, future research directions. As case GAN NextG communications, show effectively applied anomaly signal classification (e.g., user authentication) outperforming another state-of-the-art ML technique, an autoencoder.