A new prediction algorithm using covariance information in linear continuous systems

Abstract

Listen

Translate article

This paper proposes a new type of predictor, which uses the covariance information of the signal and the observation noise, for white Gaussian plus colored observation noise in linear continuous stochastic systems. This method has the advantage that one need not assume that the state-space model is known as prior necessary information. A numerical simulation example shows that the presented prediction algorithm is feasible.