Smoothed Wigner-Ville parametric modeling for the analysis of nonstationary signals

The Wigner-Ville distribution (WVD) is a useful tool in the analysis of nonstationary signals and time-varying systems due to its excellent resolution. When applied to multicomponent signals, however, it produces interference terms known as crossterms, which can hinder the correct identification of signal components. It is possible to attenuate the crossterms significantly by smoothing the WVD in time (SWVD). Another problem is the large amount of storage needed. Parametric modeling of the WVD can be used for efficient data reduction and improved frequency resolution. Crossterms, however, are still present after modeling, and are often stronger than the signal components. By estimating a parametric model of the SWVD, a clearer representation of the signal components´ evolution and a lower model order are possible. Use of the analytic signal allows each signal component to be represented by a single complex pole, thus requiring only half the usual AR model order. Further data compression is obtained by polynomial fitting to the AR parameters