Dynamic systems with random structure: an approach to the generation of nonstationary stochastic processes

The representation of nonstationary continuous-time, generally non-Gaussian processes with distinguishable states with the help of dynamic systems with random structure are considered. Switching of the partial subsystems is intended to be a Poisson point process and so the model has the Markov property and the generating system itself is described by a set of stochastic nonlinear differential equations. The statistical characteristics of the system output are analyzed. (C) 1999 The Franklin Institute. Published by Elsevier Science Ltd. All rights reserved.