Linear networks; Closed-loop identification; Identifiability; Noise power spectrum; Biological networks
Résumé :
[en] This paper considers the problem of inferring the structure and dynamics of an unknown network driven by unknown noise inputs. Equivalently we seek to identify direct causal dependencies among manifest variables only from observations of these variables. We consider
linear, time-invariant systems of minimal order and with one noise source per manifest state. It
is known that if the transfer matrix from the inputs to manifest states is minimum phase, then
this problem has a unique solution, irrespective of the network topology. Here we consider the
general case where the transfer matrix may be non-minimum phase and show that solutions are
characterized by an Algebraic Riccati Equation (ARE). Each solution to the ARE corresponds to at most one spectral factor of the output spectral density that satisfies the assumptions made.
Hence in general the problem may not have a unique solution, but all solutions can be computed
by solving an ARE and their number may be finite.
Disciplines :
Ingénierie, informatique & technologie: Multidisciplinaire, généralités & autres
Auteur, co-auteur :
Hayden, David
Yuan, Ye
GONCALVES, Jorge ; University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB)
Langue du document :
Anglais
Titre :
Network Reconstruction from Intrinsic Noise: Non-Minimum-Phase Systems
Date de publication/diffusion :
2014
Nom de la manifestation :
19th World Congress of The International Federation of Automatic Control
Lieu de la manifestation :
Cape Town, Afrique du Sud
Date de la manifestation :
August 24-29, 2014
Titre de l'ouvrage principal :
The proceedings of the The 19th World Congress of The International Federation of Automatic Control
B. D. O. Anderson (1969) "The inverse problem of stationary covariance" Journal of Statistical Physics, Vol. 1, no. 1, pp. 133-147
A. Dankers, P. M. J. Van den Hof, P. S. C. Heuberger and X. Bombois (2012) "Dynamic network identification using the direct prediction-error method" Proceedings of 51st IEEE Conference on Decision and Control
K. Glover and J. C. Willems (1974) "Parametrizations of linear dynamical systems: canonical forms and identifiability" IEEE Transactions on Automatic Control, Vol. 19, No. 6, pp. 640-645
J. Gonçalves and S. Warnick (2008) "Necessary and sufficient conditions for dynamical structure reconstruction of LTI networks" IEEE Transactions on Automatic Control, Vol. 53, pp. 1670-1674
D. Hayden, Y. Yuan and J. Gonçalves (2013) "Network Reconstruction from Intrinsic Noise" arXiv:1310.0375 [cs.SY]
D. Hayden, Y. Yuan and J. Gonçalves (2014) "Network Reconstruction from Intrinsic Noise: Minimum-Phase Systems" to appear in Proceedings of American Control Conference
P. Lancaster and L. Rodman (1995) Algebraic Riccati Equations Oxford Science Publications
D. Materassi and M. V. Salapaka (2012a) "Network reconstruction of dynamical polytrees with unobserved nodes" Proceedings of 51st IEEE Conference on Decision and Control
D. Materassi and M. V. Salapaka (2012b) "On the problem of reconstructing an unknown topology via locality properties of the weiner filter" IEEE Transactions on Automatic Control, Vol. 57, no. 7
M. Nabi-Abdolyousefi and M. Mesbahi (2012) "Network identification via node knock-outs" IEEE Transactions on Automatic Control, Vol. 57, no. 12, pp. 3214-3219
S. Shahrampour and V. M. Preciado (2013) "Topology identification of directed dynamical networks via power spectral analysis" arXiv:1308.2248 [cs.SY]
P. M. J. Van den Hof, A. Dankers, P. S. C. Heuberger and X. Bombois (2013) "Identification of dynamic models in complex networks with prediction error methods - basic methods for consistent module estimates" Automatica, Vol. 49, no. 10, pp. 2994-3006
D. C. Youla (1961) "On the factorization of rational matrices" IRE Trans. Information Theory, Vol. 7, no. 3, pp. 172-189
