2019/08/02 by Karthik R. Ramaswamy, Paul M.J. Van den Hof, Ramaswamy, Karthik R. +1 · 1 citation
Computer Science · Engineering · #93A99 #Blind Source Separation Techniques #Control Systems and Identification #FOS: Electrical engineering #Fault Detection and Control Systems #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1908.00976
openalex publication_date 2019/08/02 · openalex created_date 2022/07/28 · openalex updated_date 2026/07/28
The identification of local modules in dynamic networks with known topology\nhas recently been addressed by formulating conditions for arriving at\nconsistent estimates of the module dynamics, under the assumption of having\ndisturbances that are uncorrelated over the different nodes. The conditions\ntypically reflect the selection of a set of node signals that are taken as\npredictor inputs in a MISO identification setup. In this paper an extension is\nmade to arrive at an identification setup for the situation that process noises\non the different node signals can be correlated with each other. In this\nsituation the local module may need to be embedded in a MIMO identification\nsetup for arriving at a consistent estimate with maximum likelihood properties.\nThis requires the proper treatment of confounding variables. The result is a\nset of algorithms that, based on the given network topology and disturbance\ncorrelation structure, selects an appropriate set of node signals as predictor\ninputs and outputs in a MISO or MIMO identification setup. Three algorithms are\npresented that differ in their approach of selecting measured node signals.\nEither a maximum or a minimum number of measured node signals can be\nconsidered, as well as a preselected set of measured nodes.\n