Adaptive Chandrasekhar filter for linear discrete-time stationary stochastic systems (Q1892061)

The adaptive Chandrasekhar filter based on the SSM is developed for the adaptive estimation of the states and parameters in linear discrete time- invariant dynamical systems. Two types of models are presented to calculate the sensitivity functions with respect to the unknown parameters. One is the conventional linear innovations model and the other is the simplified model which has less computational time and storage than the innovations model. It is explained that the adaptive Chandrasekhar filtering technique is superior to the adaptive Kalman filtering technique from a computational point of view. Also, the adaptive Chandrasekhar filter is stable in numerical calculations.