Abstract:

The detection and parameter estimation of multi-component polynomial phase signals is a major difficulty in polynomial phase signal processing. Time-frequency analysis method can characterize the instantaneous frequency of polynomial phase signals by time-frequency distribution. The coefficients of the extracted instantaneous frequencies can be obtained by polynomial fitting. Finally, the parameters of the polynomial phase signal can be obtained through a simple transformation. In this study, the multi-component polynomial phase signal was separated by RPRG+ICCD to obtain the instantaneous frequency of each component, and then RANSAC was used for polynomial fitting. Finally, the parameter estimation of the polynomial phase signal was realized. The experimental results demonstrated that the proposed method could achieve the separation and parameter estimation of multi-component polynomial phase signals overlapping in the time-frequency domain.

Key words: polynomial phase signal, multi-componentssignal, parametersestimation, ridge path regroup, time frequency signal separation, random sample consensus