Orthogonal Periodic Sequences for the Identification of Functional Link Polynomial Filters

The paper introduces a novel family of deterministic periodic signals, the orthogonal periodic sequences (OPSs), that allow the perfect identification on a finite period of any functional link polynomials (FLiP) filter with the cross-correlation method. The class of FLiP filters is very broad and in...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on signal processing 2020, Vol.68, p.5308-5321
Main Authors: Carini, Alberto, Orcioni, Simone, Terenzi, Alessandro, Cecchi, Stefania
Format: Article
Language:English
Subjects:
Algebra
Computational complexity
Computational modeling
Correlation
Estimation
Functional link polynomial filters
Lips
Nonlinear filters
Nonlinear systems
ortogonal periodic sequences
Polynomials
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The paper introduces a novel family of deterministic periodic signals, the orthogonal periodic sequences (OPSs), that allow the perfect identification on a finite period of any functional link polynomials (FLiP) filter with the cross-correlation method. The class of FLiP filters is very broad and includes many popular nonlinear filters, as the well-known Volterra and the Wiener nonlinear filters. The novel sequences share many properties of the perfect periodic sequences (PPSs). As the PPSs, they allow the perfect identification of FLiP filters with the cross-correlation method. But, while PPSs exist only for orthogonal FLiP filters, the OPSs allow also the identification of non-orthogonal FLiP filters, as the Volterra filters. In OPSs, the modeled system input can be any persistently exciting sequence and can also be a quantized sequence. Moreover, OPSs can often identify FLiP filters with a sequence period and a computational complexity much smaller that PPSs. The provided experimental results, involving the identification of real devices and of a benchmark model, highlight the potentialities of the proposed OPSs in modeling unknown nonlinear systems.
ISSN:1053-587X
1941-0476
DOI:10.1109/TSP.2020.3021244