Detection of unknown nonlinear fm signals by time-frequency morphological filtering

A new energy integration detection scheme is proposed to detect an unknown nonlinear frequency modulated (FM) signal buried in strong complex additive white Gaussian noise (CAWGN). In this scheme, the Smoothed-Pseudo Wigner-Vill Distribution (SPWVD) is used to realize local coherent integration of a signal and thresholding and mathematical morphological filtering are jointly used to extract the time-frequency (TF) support region of the signal from the time-frequency distribution (TFD) image of an observation. The simulated results show that the proposed scheme is effective in strong noise background. The scheme can be applied to the target detection in the over-the-horizon radar (OTHR), such as the sky wave OTHR and the surface wave OTHR.