Abstract: As of the most important branches of artificial intelligence, deep learning (DL) has developed rapidly in recent years, and has been successfully used in many research fields. Although the DL-based algorithms offer a great opportunity for researchers to finally conquer the bottleneck problems in the field of radar signal processing, they also bring about brand-new technical challenges. In this paper, comprehensive review of the applications of DL methods is proposed, including low probability of interception and passive radar waveform recognition, automatic target recognition, radar jamming/ clutter recognition and suppression, and radar waveform and antenna array design. Recently the proposed DL-based radar waveform recognition and SAR automatic target recognition methods are summarized and analyzed in detail. The major factors limiting the performance of the DL algorithms are also examined. This work aims to provide valuable information to the scholars in this promising field of research.