The authors deal with the stochastic realization and reduction of non-minimum-phase, linear, time-invariant systems. Assuming that the input is non-Gaussian, stationary and white, they propose the use of second and higher-order output statistics to identify a finite-dimensional ARMA model. Both the input/output, and state space approaches are considered. It is shown that higher-order output cumulants are useful in estimating the AR coefficients of the minimum-phase (MP) part, and in providing a means for model order determination. Higher order cumulants are exploited to realize the all-pass part of the ARMA model.
|Original language||English (US)|
|Number of pages||6|
|Journal||Proceedings of the IEEE Conference on Decision and Control|
|State||Published - 1986|