(Almost) periodic moving average system identification using higher order cyclic-statistics

This article addresses the problem of (almost) periodic moving average (APMA) system identification. Two new normal equations relating the coefficients of an APMA system and the time-varying higher order cumulants of the measurements are established, from which two new linear algebraic algorithms are presented for system parameter estimation. In addition, a new singular value decomposition (SVD) based algorithm is proposed for determining the system order. Simulation examples are given to demonstrate the performance of these new approaches.