Research on Sea Clutter Suppression Using Sparse Dictionary Learning

Abstract

Due to the interference of the strong sea clutter, it’s difficult to detect marine weak targets and clutter attenuation is the key to improve detection performance. A sea clutter suppression method based on sparse dictionary learning was proposed to attenuate sea clutter and thus enhance the detection performance. An optimized K- singular value decomposition method (K-SVD) is firstly given to train the sea clutter learning dictionary. Secondly, we exploit an orthogonal matching pursuit method (OMP) to recover the clutter component from the measured marine signals under the clutter dictionary. Then, filter the sea clutter and extract the target component, after which the clutter-suppressed signal is under detection. We also explored the difference in sea clutter attenuation and detection performance between the traditional fixed dictionary and the learning dictionary trained in this paper. The experimental results on the CSIR public set verify the effectiveness of the algorithm.