Stealthy Adversarial Attacks on Machine Learning-Based Classifiers of Wireless Signals
Machine learning (ML) has been successfully applied to classification tasks in many domains, including computer vision, cybersecurity, and communications.Although highly accurate classifiers have been developed, research shows that these classifiers are, in general, vulnerable to adversarial machine learning (AML) attacks.In one type of AML attack,