Membership set identification with periodic inputs and orthonormal regressors

• Published in: Signal processing Vol. 86; no. 12; pp. 3778 - 3786
• Main Authors: Akçay, Hüseyin, At, Nuray
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.12.2006; Elsevier Science
• Subjects: Applied sciences; Central algorithms; Exact sciences and technology; Identification; Information theory; Information, signal and communications theory; Membership set; Orthonormal basis functions; Periodic inputs; Projection algorithms; Telecommunications and information theory; Periodic inputs; Membership set; Identification; Central algorithms; Orthonormal basis functions; Projection algorithms; Membership function; Input signal; Orthogonal function; Information processing; Algorithm; Periodic signal
• ISSN: 0165-1684; 1872-7557
• DOI: 10.1016/j.sigpro.2006.03.025

In this paper, we study the central and the projection algorithms in membership-set estimation with periodic input signals and orthonormal basis functions for the special case that the number of the estimated parameters equals the input period (or overparameterized model structures). First, we derive explicit formulae for a central algorithm and the diameter of the membership set. Then, we characterize the set of all projection algorithms.