Parameter Estimation of Multicomponent Chirp Signals via Sparse Representation

Abstract

A novel algorithm for parameter estimation of multicomponent chirp signals in complicated noise environment is proposed. By the matching pursuit (MP) algorithm, the signal is decomposed into Gabor atoms which provide sparse information that represents the signal time-frequency signature. The Hough transform (HT) is then directly used to estimate the parameter of chirp components without computing the time-frequency distribution. Simulation results show that this algorithm is capable of estimating parameters of multicomponent chirp signals even in the presence of strong intended interference and colored noise.