Adversarial Machine Learning in Wireless Communications using RF Data: A Review

Machine learning provides effective means to learn from spectrum data and solve complex tasks involved in wireless communications. Supported by recent advances in computational resources and algorithmic designs, deep learning has found success in performing various wireless communication tasks such as signal recognition and spectrum sensing. However, machine learning in general and deep learning in particular has recently been found vulnerable to manipulations in training and test times giving rise to a field of study called Adversarial Machine Learning (AML). Although AML has been extensively studied in other data domains such as computer vision and natural language processing, research for AML in the wireless communications domain is in its early stage. This paper presents a comprehensive review of the latest research efforts focused on AML in wireless communications while accounting for the unique characteristics of wireless systems. First, the necessary background on the various types of AML attacks is provided. Then, a holistic survey of the works developing the AML attacks and the corresponding defense mechanisms in the wireless domain is presented. Finally, recent research trends are identified and the future outlook for AML as a new attack surface for wireless communications is described.