Linear Optimal Kalman Predictor for Descriptor System With Packet Dropouts and Colored Measurement Noises

This paper addresses the optimal linear estimation problem for a descriptor system with missing measurements, packet dropouts and colored measurement noises. By singular value decomposition, fictitious noise compensation and augmented state approach, the original system can be transformed into two reduced-order subsystems with augmented correlated noises. Then the problem of optimal Kalman predicting is transformed into prediction problem for augmented systems. Based on Kalman filtering theory, the steady-state Kalman predictor of the original descriptor system is presented. The effectiveness of the proposed algorithm is verified by a numerical simulation example.