Parameter estimation of multicomponent polynomial-phase signals by intersection of signal subspaces

The aim of this work is the parameter estimation of polynomial-phase signals (PPSs) embedded in white noise. The proposed estimation algorithm is a generalization of the method based on the polynomial-phase transform (PPT), able to solve an ambiguity problem appearing when applying the PPT to PPSs having the same highest order phase coefficients. The proposed approach is based on the intersection of two (or more) signal subspaces sharing only the useful components but not the undesired spurious harmonics. Different signal subspaces are obtained by exploiting a multilag definition of the PPT.