Detection of marine target with quadratic modulated frequency micromotion signature via Morphological Component Analysis

Using the sparse property of the signal from a marine target with quadratic modulated frequency (QFM) micromotion signature, the detection in sparse domain is studied. An algorithm for separation of sea clutter and micro-Doppler signal is proposed based on the Morphological Component Analysis (MCA). At first, the signal model of a marine target with micromotion is established, which can be modelled as a QFM signal. Then, the MCA principle and application to the micro-Doppler signal are introduced, which is in fact a minimization problem. The key step is the choice of dictionary. We propose an algorithm for dictionary construction based on the phase differentiation and the property of micro-Doppler (m-D) signal. Finally, experiment using real radar data indicate that the proposed method can achieve excellent integration and detection performance.