Optimal filtering for discrete time-varying systems with multiple time-delay measurements

This paper aims to give the efficient approach to the discrete-time systems with instantaneous and multiple time-delays measurements. Without resorting to the traditional augmentation and standard Kalman filtering formulation, a new approach termed as reorganized innovation analysis is developed in the paper. The improved performance is clearly demonstrated through analytical results and simulation experiments with multiple time-delayed measurements. More importantly, the approach presented in the paper forms the basis of solving the more complicated problems such as H_∞ fixed-lag smoothing, multiple-step ahead prediction and H_∞ control with control input multiple delays.