Modeling and estimation of a class of dynamic multiscale system subject to colored noise

In this paper, modeling and estimation of a class of dynamic multiscale system subject to colored state equation noise and measurement equation noise is proposed. The state equation is whitened firstly and then the measurement equation. The state space projection equation is used to link the scales, then a new system model is built. The new model is in a form suitable for the application of the Kalman filter equations. Haar-wavelet-based model and estimation algorithm are given. Monte Carlo simulation results demonstrate that the proposed algorithm is effective and powerful in this kind of multiscale estimation problem.