Title :
Identifiability of minimal, stable, rational, causal systems using second-order output cyclic spectra
Author_Institution :
Dept. of Electr. & Syst. Eng., Connecticut Univ., Storrs, CT, USA
fDate :
5/1/1995 12:00:00 AM
Abstract :
In this paper, the (blind) identifiability of systems using the second-order output cyclic spectra is considered. It is shown that the class of minimal, stable, rational causal (MSRC) systems is not identifiable, and the identification is limited by a nontrivial ambiguity associated with the so-called cyclic all-pass factor. The largest subclass of MSRC systems that ensures the unique identification is shown to be the set of MSRC systems that do not have special uniformly spaced maximum phase zeros
Keywords :
identification; stability; blind identifiability; cyclic all-pass factor; minimal stable rational causal systems; nontrivial ambiguity; second-order output cyclic spectra; unique identification; Communication system control; Data processing; Estimation error; Finite impulse response filter; Higher order statistics; Information retrieval; Nearest neighbor searches; Signal processing; Target tracking; Transfer functions;
Journal_Title :
Automatic Control, IEEE Transactions on