DNCNet: Deep Radar Signal Denoising and Recognition

Deep learning with its rapid development and advancement has achieved unparalleled performance in many areas like computer vision as well cognitive radio signal recognition. However, the of most deep neural networks would suffer from degradation data mismatch scenario, e.g., test dataset a related but nonidentical distribution training dataset. Considering noise corruption, classifier’s accuracy might drop sharply when it is tested on much lower signal-to-noise ratio compared to To address this dilemma, work, we propose an efficient denoising classification network (DNCNet) for radar signals. The DNCNet consists subnetworks. First, detection synthetic mechanism designed generate pairwise clean noisy train subnetwork. Then, two-phase procedure proposed subnetwork first phase strengthen mapping between results perceptual representation second. Experiments benchmark datasets validate excellent against state-of-the-art methods terms both restoration quality accuracy.