The Influence of Data Loss on Diagnostics of Complex System Dynamics

Abstract

We consider the problem of quantitative description of the dynamics of complex systems using experimental data records containing various breakdown regions related to artifacts, violations of the operation conditions, or monitoring of equipment malfunctions. Elimination (rejection) of these regions can lead to modification of the signal structure, which makes it necessary to assess the influence of this procedure on the quantitative characteristics that are to be determined. An example of multiscale analysis of cerebral blood flow is presented, which shows that the correct diagnostics of the regime of system functioning may be possible even with extreme loss of data.