Abstract:

The echo scattered by a slow-moving weak target is much weaker than the sea clutter, and they are very close in the time or frequency domain,which presents a difficult task for the marine surface surveillance radar. To detect a slow-moving weak target, considering the fact that both the target echo and sea clutter are non-stationary, we propose a slow-moving target detector based on the convex hull of the time-frequency concentration and the lasting time of the decomposed components, which are obtained by the proposed time-frequency decomposing method. The proposed detector can detect the target signal with much weaker energy than that of the sea clutter, with the two being very close in the time-frequency domain. It is evaluated by the IPIX radar datasets with a simulated target or a real target. Results demonstrate that the proposed method can not only detect a slow-moving weak target in high accuracy, but also provide the instantaneous state of the target.

Key words: time-frequency distribution, time-frequency decomposition, sea surface echo, target detection