Nonminimum-phase complex Fourier series based model for statistical signal processing

In this paper, a parametric complex Fourier series based model (FSBM), an extension of the real FSBM proposed by Chi, for or as an approximation to an arbitrary nonminimum-phase complex linear time-invariant (LTI) system is proposed for statistical signal processing applications where signals and LTI systems of interest are complex. Based on the proposed complex FSBM, a complex linear prediction error (LPE) filter is presented along with the Cramer Rao (CR) bound for the case of finite Gaussian measurements. Then an MP-AP algorithm is presented for the estimation of the complex FSBM parameters followed by some simulation results for channel identification and equalization in communications.