Research on signals extrapolation in the time domain based on wavelet analysis

Characteristic of modulation spectrum is important for radar targets classification and recognition, while signals with high occupation ratio cause low spectral resolution. For this case, the authors present a method for splicing sub‐band based on the Mallat algorithm. This method extracts the low‐frequency parts in the different periods of the signals, and using the state‐space method to extrapolate the signal between pulse sequence. This method sacrifices spectrum width to improve the local spectral resolution in the case of less sample data. The simulation results demonstrate that the method is effective.