Spatial analysis using new properties of the cross-spectral matrix

Modelling of Non Stationary Systems Based on a Dynamical Decision Space

“…Several studies (Mercère et al, 2004;Mercère et al, 2005) have enhanced some theoretical and practical benefits of the adaptation of a particular array signal processing technique: the propagator method (Munier and Delisle, 1991). The main advantage of this approach lies in the use of a linear operator and unconstrained and unapproximated quadratic criteria which lead to easy recursive least squares algorithms.…”

“…The complement Γ Γ Γ f 2 of Γ Γ Γ f 1 can be expressed as a linear combination of these n x rows. There is a unique linear operator P f ∈ R n x ×(n y f −n x ) , named propagator (Munier and Delisle, 1991), such that…”

Convergence Analysis of Instrumental Variable Recursive Subspace Identification Algorithms

“…A number of methods have been developed in the literature in order to avoid the computation of a complete SVD in this step of recursive subspace identification [9,14,10]. In this paper, the focus is on algorithms based on the propagator concept [13].…”

“…Then, the complement Γ f2 of Γ f1 can be expressed as a linear combination of these n x rows. So, there is a unique linear operator P f ∈ R nx×(nyf −nx) , named propagator [13], such that…”

“…• The second one rests on the adaptation of another array signal processing technique: the propagator method [13]. The advantage of this approach over the previous conception lies in the use of a linear operator and quadratic criteria which lead to recursive least squares implementations for the algorithms [11,10,12].…”